hadoop.hdfs.configuration.version
1
version of this configuration file
dfs.namenode.rpc-address
RPC address that handles all clients requests. In the case of HA/Federation where multiple namenodes exist,
the name service id is added to the name e.g. dfs.namenode.rpc-address.ns1
dfs.namenode.rpc-address.EXAMPLENAMESERVICE
The value of this property will take the form of nn-host1:rpc-port. The NameNode's default RPC port is 8020.
dfs.namenode.rpc-bind-host
The actual address the RPC server will bind to. If this optional address is
set, it overrides only the hostname portion of dfs.namenode.rpc-address.
It can also be specified per name node or name service for HA/Federation.
This is useful for making the name node listen on all interfaces by
setting it to 0.0.0.0.
dfs.namenode.servicerpc-address
RPC address for HDFS Services communication. BackupNode, Datanodes and all other services should be
connecting to this address if it is configured. In the case of HA/Federation where multiple namenodes exist,
the name service id is added to the name e.g. dfs.namenode.servicerpc-address.ns1
dfs.namenode.rpc-address.EXAMPLENAMESERVICE
The value of this property will take the form of nn-host1:rpc-port.
If the value of this property is unset the value of dfs.namenode.rpc-address will be used as the default.
dfs.namenode.servicerpc-bind-host
The actual address the service RPC server will bind to. If this optional address is
set, it overrides only the hostname portion of dfs.namenode.servicerpc-address.
It can also be specified per name node or name service for HA/Federation.
This is useful for making the name node listen on all interfaces by
setting it to 0.0.0.0.
dfs.namenode.lifeline.rpc-address
NameNode RPC lifeline address. This is an optional separate RPC address
that can be used to isolate health checks and liveness to protect against
resource exhaustion in the main RPC handler pool. In the case of
HA/Federation where multiple NameNodes exist, the name service ID is added
to the name e.g. dfs.namenode.lifeline.rpc-address.ns1. The value of this
property will take the form of nn-host1:rpc-port. If this property is not
defined, then the NameNode will not start a lifeline RPC server. By
default, the property is not defined.
dfs.namenode.lifeline.rpc-bind-host
The actual address the lifeline RPC server will bind to. If this optional
address is set, it overrides only the hostname portion of
dfs.namenode.lifeline.rpc-address. It can also be specified per name node
or name service for HA/Federation. This is useful for making the name node
listen on all interfaces by setting it to 0.0.0.0.
dfs.namenode.secondary.http-address
0.0.0.0:9868
The secondary namenode http server address and port.
dfs.namenode.secondary.https-address
0.0.0.0:9869
The secondary namenode HTTPS server address and port.
dfs.datanode.address
0.0.0.0:9866
The datanode server address and port for data transfer.
dfs.datanode.http.address
0.0.0.0:9864
The datanode http server address and port.
dfs.datanode.ipc.address
0.0.0.0:9867
The datanode ipc server address and port.
dfs.datanode.http.internal-proxy.port
0
The datanode's internal web proxy port.
By default it selects a random port available in runtime.
dfs.datanode.handler.count
10
The number of Datanode RPC server threads that listen to
requests from client.
dfs.namenode.http-address
0.0.0.0:9870
The address and the base port where the dfs namenode web ui will listen on.
dfs.namenode.http-bind-host
The actual address the HTTP server will bind to. If this optional address
is set, it overrides only the hostname portion of dfs.namenode.http-address.
It can also be specified per name node or name service for HA/Federation.
This is useful for making the name node HTTP server listen on all
interfaces by setting it to 0.0.0.0.
dfs.namenode.heartbeat.recheck-interval
300000
This time decides the interval to check for expired datanodes.
With this value and dfs.heartbeat.interval, the interval of
deciding the datanode is stale or not is also calculated.
The unit of this configuration is millisecond.
dfs.http.policy
HTTP_ONLY
Decide if HTTPS(SSL) is supported on HDFS
This configures the HTTP endpoint for HDFS daemons:
The following values are supported:
- HTTP_ONLY : Service is provided only on http
- HTTPS_ONLY : Service is provided only on https
- HTTP_AND_HTTPS : Service is provided both on http and https
dfs.client.https.need-auth
false
Whether SSL client certificate authentication is required
dfs.client.cached.conn.retry
3
The number of times the HDFS client will pull a socket from the
cache. Once this number is exceeded, the client will try to create a new
socket.
dfs.https.server.keystore.resource
ssl-server.xml
Resource file from which ssl server keystore
information will be extracted
dfs.client.https.keystore.resource
ssl-client.xml
Resource file from which ssl client keystore
information will be extracted
dfs.datanode.https.address
0.0.0.0:9865
The datanode secure http server address and port.
dfs.namenode.https-address
0.0.0.0:9871
The namenode secure http server address and port.
dfs.namenode.https-bind-host
The actual address the HTTPS server will bind to. If this optional address
is set, it overrides only the hostname portion of dfs.namenode.https-address.
It can also be specified per name node or name service for HA/Federation.
This is useful for making the name node HTTPS server listen on all
interfaces by setting it to 0.0.0.0.
dfs.datanode.dns.interface
default
The name of the Network Interface from which a data node should
report its IP address. e.g. eth2. This setting may be required for some
multi-homed nodes where the DataNodes are assigned multiple hostnames
and it is desirable for the DataNodes to use a non-default hostname.
Prefer using hadoop.security.dns.interface over
dfs.datanode.dns.interface.
dfs.datanode.dns.nameserver
default
The host name or IP address of the name server (DNS) which a DataNode
should use to determine its own host name.
Prefer using hadoop.security.dns.nameserver over
dfs.datanode.dns.nameserver.
dfs.namenode.backup.address
0.0.0.0:50100
The backup node server address and port.
If the port is 0 then the server will start on a free port.
dfs.namenode.backup.http-address
0.0.0.0:50105
The backup node http server address and port.
If the port is 0 then the server will start on a free port.
dfs.namenode.redundancy.considerLoad
true
Decide if chooseTarget considers the target's load or not when write.
Turn on by default.
dfs.namenode.redundancy.considerLoadByStorageType
false
Decide if chooseTarget considers the target's load with respect to the
storage type. Typically to be used when datanodes contain homogenous
storage types. Irrelevent if dfs.namenode.redundancy.considerLoad is
false.
dfs.namenode.redundancy.considerLoad.factor
2.0
The factor by which a node's load can exceed the average
before being rejected for writes, only if considerLoad is true.
dfs.namenode.redundancy.considerLoadByVolume
false
Decide if chooseTarget considers the target's volume load or
not.
dfs.namenode.read.considerLoad
false
Decide if sort block locations considers the target's load or not when read.
Turn off by default.
It is not possible to enable this feature along with dfs.namenode.read.considerStorageType as only one sort can be
enabled at a time.
dfs.namenode.read.considerStorageType
false
Decide if sort block locations considers the target's storage type or not when read. Any locations with the same
network distance are sorted in order of the storage speed, fastest first (RAM, SSD, Disk, Archive). This is
disabled by default, and the locations will be ordered randomly.
It is not possible to enable this feature along with dfs.namenode.read.considerLoad as only one sort can be
enabled at a time.
dfs.datanode.httpserver.filter.handlers
org.apache.hadoop.hdfs.server.datanode.web.RestCsrfPreventionFilterHandler
Comma separated list of Netty servlet-style filter handlers to inject into the Datanode WebHDFS I/O path
dfs.default.chunk.view.size
32768
The number of bytes to view for a file on the browser.
dfs.datanode.du.reserved.calculator
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.ReservedSpaceCalculator$ReservedSpaceCalculatorAbsolute
Determines the class of ReservedSpaceCalculator to be used for
calculating disk space reservedfor non-HDFS data. The default calculator is
ReservedSpaceCalculatorAbsolute which will use dfs.datanode.du.reserved
for a static reserved number of bytes. ReservedSpaceCalculatorPercentage
will use dfs.datanode.du.reserved.pct to calculate the reserved number
of bytes based on the size of the storage. ReservedSpaceCalculatorConservative and
ReservedSpaceCalculatorAggressive will use their combination, Conservative will use
maximum, Aggressive minimum. For more details see ReservedSpaceCalculator.
dfs.datanode.du.reserved
0
Reserved space in bytes per volume. Always leave this much space free for non dfs use.
Specific storage type based reservation is also supported. The property can be followed with
corresponding storage types ([ssd]/[disk]/[archive]/[ram_disk]/[nvdimm]) for cluster with heterogeneous storage.
For example, reserved space for RAM_DISK storage can be configured using property
'dfs.datanode.du.reserved.ram_disk'. If specific storage type reservation is not configured
then dfs.datanode.du.reserved will be used. Support multiple size unit suffix(case insensitive),
as described in dfs.blocksize.
Note: In case of using tune2fs to set reserved-blocks-percentage, or other filesystem tools,
then you can possibly run into out of disk errors because hadoop will not check those
external tool configurations.
dfs.datanode.du.reserved.pct
0
Reserved space in percentage. Read dfs.datanode.du.reserved.calculator to see
when this takes effect. The actual number of bytes reserved will be calculated by using the
total capacity of the data directory in question. Specific storage type based reservation
is also supported. The property can be followed with corresponding storage types
([ssd]/[disk]/[archive]/[ram_disk]/[nvdimm]) for cluster with heterogeneous storage.
For example, reserved percentage space for RAM_DISK storage can be configured using property
'dfs.datanode.du.reserved.pct.ram_disk'. If specific storage type reservation is not configured
then dfs.datanode.du.reserved.pct will be used.
dfs.namenode.name.dir
file://${hadoop.tmp.dir}/dfs/name
Determines where on the local filesystem the DFS name node
should store the name table(fsimage). If this is a comma-delimited list
of directories then the name table is replicated in all of the
directories, for redundancy.
dfs.namenode.name.dir.restore
false
Set to true to enable NameNode to attempt recovering a
previously failed dfs.namenode.name.dir. When enabled, a recovery of any
failed directory is attempted during checkpoint.
dfs.namenode.fs-limits.max-component-length
255
Defines the maximum number of bytes in UTF-8 encoding in each
component of a path. A value of 0 will disable the check. Support
multiple size unit suffix(case insensitive), as described in dfs.blocksize.
dfs.namenode.fs-limits.max-directory-items
1048576
Defines the maximum number of items that a directory may
contain. Cannot set the property to a value less than 1 or more than
6400000.
dfs.namenode.fs-limits.min-block-size
1048576
Minimum block size in bytes, enforced by the Namenode at create
time. This prevents the accidental creation of files with tiny block
sizes (and thus many blocks), which can degrade performance. Support multiple
size unit suffix(case insensitive), as described in dfs.blocksize.
dfs.namenode.fs-limits.max-blocks-per-file
10000
Maximum number of blocks per file, enforced by the Namenode on
write. This prevents the creation of extremely large files which can
degrade performance.
dfs.namenode.edits.dir
${dfs.namenode.name.dir}
Determines where on the local filesystem the DFS name node
should store the transaction (edits) file. If this is a comma-delimited list
of directories then the transaction file is replicated in all of the
directories, for redundancy. Default value is same as dfs.namenode.name.dir
dfs.namenode.edits.dir.required
This should be a subset of dfs.namenode.edits.dir,
to ensure that the transaction (edits) file
in these places is always up-to-date.
dfs.namenode.shared.edits.dir
A directory on shared storage between the multiple namenodes
in an HA cluster. This directory will be written by the active and read
by the standby in order to keep the namespaces synchronized. This directory
does not need to be listed in dfs.namenode.edits.dir above. It should be
left empty in a non-HA cluster.
dfs.namenode.edits.journal-plugin.qjournal
org.apache.hadoop.hdfs.qjournal.client.QuorumJournalManager
dfs.namenode.edits.qjournals.resolution-enabled
false
Determines if the given qjournals address is a domain name which needs to
be resolved.
This is used by namenode to resolve qjournals.
dfs.namenode.edits.qjournals.resolver.impl
Qjournals resolver implementation used by namenode.
Effective with dfs.namenode.edits.qjournals.resolution-enabled on.
dfs.permissions.enabled
true
If "true", enable permission checking in HDFS.
If "false", permission checking is turned off,
but all other behavior is unchanged.
Switching from one parameter value to the other does not change the mode,
owner or group of files or directories.
dfs.permissions.ContentSummary.subAccess
false
If "true", the ContentSummary permission checking will use subAccess.
If "false", the ContentSummary permission checking will NOT use subAccess.
subAccess means using recursion to check the access of all descendants.
dfs.permissions.superusergroup
supergroup
The name of the group of super-users.
The value should be a single group name.
dfs.cluster.administrators
ACL for the admins, this configuration is used to control
who can access the default servlets in the namenode, etc. The value
should be a comma separated list of users and groups. The user list
comes first and is separated by a space followed by the group list,
e.g. "user1,user2 group1,group2". Both users and groups are optional,
so "user1", " group1", "", "user1 group1", "user1,user2 group1,group2"
are all valid (note the leading space in " group1"). '*' grants access
to all users and groups, e.g. '*', '* ' and ' *' are all valid.
dfs.namenode.ip-proxy-users
A comma separated list of user names that are allowed by the
NameNode to specify a different client IP address in the caller context.
This is used by Router-Based Federation (RBF) to provide the actual client's
IP address to the NameNode, which is critical to preserve data locality when
using RBF. If you are using RBF, add the user that runs the routers.
dfs.namenode.acls.enabled
true
Set to true to enable support for HDFS ACLs (Access Control Lists). By
default, ACLs are enabled. When ACLs are disabled, the NameNode rejects
all RPCs related to setting or getting ACLs.
dfs.namenode.posix.acl.inheritance.enabled
true
Set to true to enable POSIX style ACL inheritance. When it is enabled
and the create request comes from a compatible client, the NameNode
will apply default ACLs from the parent directory to the create mode
and ignore the client umask. If no default ACL found, it will apply the
client umask.
dfs.namenode.lazypersist.file.scrub.interval.sec
300
The NameNode periodically scans the namespace for LazyPersist files with
missing blocks and unlinks them from the namespace. This configuration key
controls the interval between successive scans. If this value is set to 0,
the file scrubber is disabled.
dfs.block.access.token.enable
false
If "true", access tokens are used as capabilities for accessing datanodes.
If "false", no access tokens are checked on accessing datanodes.
dfs.block.access.key.update.interval
600
Interval in minutes at which namenode updates its access keys.
dfs.block.access.token.lifetime
600
The lifetime of access tokens in minutes.
dfs.block.access.token.protobuf.enable
false
If "true", block tokens are written using Protocol Buffers.
If "false", block tokens are written using Legacy format.
dfs.datanode.data.dir
file://${hadoop.tmp.dir}/dfs/data
Determines where on the local filesystem an DFS data node
should store its blocks. If this is a comma-delimited
list of directories, then data will be stored in all named
directories, typically on different devices. The directories should be tagged
with corresponding storage types ([SSD]/[DISK]/[ARCHIVE]/[RAM_DISK]/[NVDIMM]) for HDFS
storage policies. The default storage type will be DISK if the directory does
not have a storage type tagged explicitly. Directories that do not exist will
be created if local filesystem permission allows.
dfs.datanode.data.dir.perm
700
Permissions for the directories on on the local filesystem where
the DFS data node store its blocks. The permissions can either be octal or
symbolic.
dfs.replication
3
Default block replication.
The actual number of replications can be specified when the file is created.
The default is used if replication is not specified in create time.
dfs.replication.max
512
Maximal block replication.
dfs.namenode.replication.min
1
Minimal block replication.
dfs.namenode.maintenance.replication.min
1
Minimal live block replication in existence of maintenance mode.
dfs.namenode.safemode.replication.min
a separate minimum replication factor for calculating safe block count.
This is an expert level setting.
Setting this lower than the dfs.namenode.replication.min
is not recommend and/or dangerous for production setups.
When it's not set it takes value from dfs.namenode.replication.min
dfs.namenode.max-corrupt-file-blocks-returned
100
The maximum number of corrupt file blocks listed by NameNode Web UI,
JMX and other client request.
dfs.blocksize
134217728
The default block size for new files, in bytes.
You can use the following suffix (case insensitive):
k(kilo), m(mega), g(giga), t(tera), p(peta), e(exa) to specify the size (such as 128k, 512m, 1g, etc.),
Or provide complete size in bytes (such as 134217728 for 128 MB).
dfs.client.block.write.retries
3
The number of retries for writing blocks to the data nodes,
before we signal failure to the application.
dfs.client.block.write.replace-datanode-on-failure.enable
true
If there is a datanode/network failure in the write pipeline,
DFSClient will try to remove the failed datanode from the pipeline
and then continue writing with the remaining datanodes. As a result,
the number of datanodes in the pipeline is decreased. The feature is
to add new datanodes to the pipeline.
This is a site-wide property to enable/disable the feature.
When the cluster size is extremely small, e.g. 3 nodes or less, cluster
administrators may want to set the policy to NEVER in the default
configuration file or disable this feature. Otherwise, users may
experience an unusually high rate of pipeline failures since it is
impossible to find new datanodes for replacement.
See also dfs.client.block.write.replace-datanode-on-failure.policy
dfs.client.block.write.replace-datanode-on-failure.policy
DEFAULT
This property is used only if the value of
dfs.client.block.write.replace-datanode-on-failure.enable is true.
ALWAYS: always add a new datanode when an existing datanode is removed.
NEVER: never add a new datanode.
DEFAULT:
Let r be the replication number.
Let n be the number of existing datanodes.
Add a new datanode only if r is greater than or equal to 3 and either
(1) floor(r/2) is greater than or equal to n; or
(2) r is greater than n and the block is hflushed/appended.
dfs.client.block.write.replace-datanode-on-failure.best-effort
false
This property is used only if the value of
dfs.client.block.write.replace-datanode-on-failure.enable is true.
Best effort means that the client will try to replace a failed datanode
in write pipeline (provided that the policy is satisfied), however, it
continues the write operation in case that the datanode replacement also
fails.
Suppose the datanode replacement fails.
false: An exception should be thrown so that the write will fail.
true : The write should be resumed with the remaining datandoes.
Note that setting this property to true allows writing to a pipeline
with a smaller number of datanodes. As a result, it increases the
probability of data loss.
dfs.client.block.write.replace-datanode-on-failure.min-replication
0
The minimum number of replications that are needed to not to fail
the write pipeline if new datanodes can not be found to replace
failed datanodes (could be due to network failure) in the write pipeline.
If the number of the remaining datanodes in the write pipeline is greater
than or equal to this property value, continue writing to the remaining nodes.
Otherwise throw exception.
If this is set to 0, an exception will be thrown, when a replacement
can not be found.
See also dfs.client.block.write.replace-datanode-on-failure.policy
dfs.blockreport.intervalMsec
21600000
Determines block reporting interval in milliseconds.
dfs.blockreport.initialDelay
0
Delay for first block report in seconds. Support multiple time unit
suffix(case insensitive), as described in dfs.heartbeat.interval.If
no time unit is specified then seconds is assumed
dfs.blockreport.split.threshold
1000000
If the number of blocks on the DataNode is below this
threshold then it will send block reports for all Storage Directories
in a single message.
If the number of blocks exceeds this threshold then the DataNode will
send block reports for each Storage Directory in separate messages.
Set to zero to always split.
dfs.namenode.max.full.block.report.leases
6
The maximum number of leases for full block reports that the
NameNode will issue at any given time. This prevents the NameNode from
being flooded with full block reports that use up all the RPC handler
threads. This number should never be more than the number of RPC handler
threads or less than 1.
dfs.namenode.full.block.report.lease.length.ms
300000
The number of milliseconds that the NameNode will wait before invalidating
a full block report lease. This prevents a crashed DataNode from
permanently using up a full block report lease.
dfs.datanode.directoryscan.interval
21600
Interval in seconds for Datanode to scan data directories and
reconcile the difference between blocks in memory and on the disk.
Support multiple time unit suffix(case insensitive), as described
in dfs.heartbeat.interval.If no time unit is specified then seconds
is assumed.
dfs.datanode.directoryscan.threads
1
How many threads should the threadpool used to compile reports
for volumes in parallel have.
dfs.datanode.directoryscan.throttle.limit.ms.per.sec
1000
The report compilation threads are limited to only running for
a given number of milliseconds per second, as configured by the
property. The limit is taken per thread, not in aggregate, e.g. setting
a limit of 100ms for 4 compiler threads will result in each thread being
limited to 100ms, not 25ms.
Note that the throttle does not interrupt the report compiler threads, so the
actual running time of the threads per second will typically be somewhat
higher than the throttle limit, usually by no more than 20%.
Setting this limit to 1000 disables compiler thread throttling. Only
values between 1 and 1000 are valid. Setting an invalid value will result
in the throttle being disabled and an error message being logged. 1000 is
the default setting.
dfs.datanode.reconcile.blocks.batch.size
1000
Setting this to define reconcile batch size.
dfs.datanode.reconcile.blocks.batch.interval
2000
Setting this to define interval between batches.
dfs.heartbeat.interval
3
Determines datanode heartbeat interval in seconds.
Can use the following suffix (case insensitive):
ms(millis), s(sec), m(min), h(hour), d(day)
to specify the time (such as 2s, 2m, 1h, etc.).
Or provide complete number in seconds (such as 30 for 30 seconds).
If no time unit is specified then seconds is assumed.
dfs.datanode.lifeline.interval.seconds
Sets the interval in seconds between sending DataNode Lifeline Protocol
messages from the DataNode to the NameNode. The value must be greater than
the value of dfs.heartbeat.interval. If this property is not defined, then
the default behavior is to calculate the interval as 3x the value of
dfs.heartbeat.interval. Note that normal heartbeat processing may cause the
DataNode to postpone sending lifeline messages if they are not required.
Under normal operations with speedy heartbeat processing, it is possible
that no lifeline messages will need to be sent at all. This property has no
effect if dfs.namenode.lifeline.rpc-address is not defined.
dfs.namenode.handler.count
10
The number of Namenode RPC server threads that listen to
requests from clients.
If dfs.namenode.servicerpc-address is not configured then
Namenode RPC server threads listen to requests from all nodes.
dfs.namenode.service.handler.count
10
The number of Namenode RPC server threads that listen to
requests from DataNodes and from all other non-client nodes.
dfs.namenode.service.handler.count will be valid only if
dfs.namenode.servicerpc-address is configured.
dfs.namenode.lifeline.handler.ratio
0.10
A ratio applied to the value of dfs.namenode.handler.count, which then
provides the number of RPC server threads the NameNode runs for handling the
lifeline RPC server. For example, if dfs.namenode.handler.count is 100, and
dfs.namenode.lifeline.handler.factor is 0.10, then the NameNode starts
100 * 0.10 = 10 threads for handling the lifeline RPC server. It is common
to tune the value of dfs.namenode.handler.count as a function of the number
of DataNodes in a cluster. Using this property allows for the lifeline RPC
server handler threads to be tuned automatically without needing to touch a
separate property. Lifeline message processing is lightweight, so it is
expected to require many fewer threads than the main NameNode RPC server.
This property is not used if dfs.namenode.lifeline.handler.count is defined,
which sets an absolute thread count. This property has no effect if
dfs.namenode.lifeline.rpc-address is not defined.
dfs.namenode.lifeline.handler.count
Sets an absolute number of RPC server threads the NameNode runs for handling
the DataNode Lifeline Protocol and HA health check requests from ZKFC. If
this property is defined, then it overrides the behavior of
dfs.namenode.lifeline.handler.ratio. By default, it is not defined. This
property has no effect if dfs.namenode.lifeline.rpc-address is not defined.
dfs.namenode.safemode.threshold-pct
0.999f
Specifies the percentage of blocks that should satisfy
the minimal replication requirement defined by dfs.namenode.replication.min.
Values less than or equal to 0 mean not to wait for any particular
percentage of blocks before exiting safemode.
Values greater than 1 will make safe mode permanent.
dfs.namenode.safemode.min.datanodes
0
Specifies the number of datanodes that must be considered alive
before the name node exits safemode.
Values less than or equal to 0 mean not to take the number of live
datanodes into account when deciding whether to remain in safe mode
during startup.
Values greater than the number of datanodes in the cluster
will make safe mode permanent.
dfs.namenode.safemode.recheck.interval
1000
Interval in msec for checking safe mode.
dfs.namenode.safemode.extension
30000
Determines extension of safe mode in milliseconds after the threshold level
is reached. Support multiple time unit suffix (case insensitive), as
described in dfs.heartbeat.interval.
dfs.namenode.resource.check.interval
5000
The interval in milliseconds at which the NameNode resource checker runs.
The checker calculates the number of the NameNode storage volumes whose
available spaces are more than dfs.namenode.resource.du.reserved, and
enters safemode if the number becomes lower than the minimum value
specified by dfs.namenode.resource.checked.volumes.minimum.
dfs.namenode.resource.du.reserved
104857600
The amount of space to reserve/require for a NameNode storage directory
in bytes. The default is 100MB. Support multiple size unit
suffix(case insensitive), as described in dfs.blocksize.
dfs.namenode.resource.checked.volumes
A list of local directories for the NameNode resource checker to check in
addition to the local edits directories.
dfs.namenode.resource.checked.volumes.minimum
1
The minimum number of redundant NameNode storage volumes required.
dfs.datanode.balance.bandwidthPerSec
100m
Specifies the maximum amount of bandwidth that each datanode
can utilize for the balancing purpose in term of
the number of bytes per second. You can use the following
suffix (case insensitive):
k(kilo), m(mega), g(giga), t(tera), p(peta), e(exa)to specify the size
(such as 128k, 512m, 1g, etc.).
Or provide complete size in bytes (such as 134217728 for 128 MB).
dfs.hosts
Names a file that contains a list of hosts that are
permitted to connect to the namenode. The full pathname of the file
must be specified. If the value is empty, all hosts are
permitted.
dfs.hosts.exclude
Names a file that contains a list of hosts that are
not permitted to connect to the namenode. The full pathname of the
file must be specified. If the value is empty, no hosts are
excluded.
dfs.namenode.max.objects
0
The maximum number of files, directories and blocks
dfs supports. A value of zero indicates no limit to the number
of objects that dfs supports.
dfs.namenode.datanode.registration.ip-hostname-check
true
If true (the default), then the namenode requires that a connecting
datanode's address must be resolved to a hostname. If necessary, a reverse
DNS lookup is performed. All attempts to register a datanode from an
unresolvable address are rejected.
It is recommended that this setting be left on to prevent accidental
registration of datanodes listed by hostname in the excludes file during a
DNS outage. Only set this to false in environments where there is no
infrastructure to support reverse DNS lookup.
dfs.namenode.decommission.interval
30
Namenode periodicity in seconds to check if
decommission or maintenance is complete. Support multiple time unit
suffix(case insensitive), as described in dfs.heartbeat.interval.
If no time unit is specified then seconds is assumed.
dfs.namenode.decommission.blocks.per.interval
500000
The approximate number of blocks to process per decommission
or maintenance interval, as defined in dfs.namenode.decommission.interval.
dfs.namenode.decommission.max.concurrent.tracked.nodes
100
The maximum number of decommission-in-progress or
entering-maintenance datanodes nodes that will be tracked at one time by
the namenode. Tracking these datanode consumes additional NN memory
proportional to the number of blocks on the datnode. Having a conservative
limit reduces the potential impact of decommissioning or maintenance of
a large number of nodes at once.
A value of 0 means no limit will be enforced.
dfs.namenode.decommission.monitor.class
org.apache.hadoop.hdfs.server.blockmanagement.DatanodeAdminDefaultMonitor
Determines the implementation used for the decommission manager. The only
valid options are:
org.apache.hadoop.hdfs.server.blockmanagement.DatanodeAdminDefaultMonitor
org.apache.hadoop.hdfs.server.blockmanagement.DatanodeAdminBackoffMonitor
dfs.namenode.decommission.backoff.monitor.pending.limit
10000
When the Backoff monitor is enabled, determines the maximum number of blocks
related to decommission and maintenance operations that can be loaded
into the replication queue at any given time. Every
dfs.namenode.decommission.interval seconds, the list is checked to see if
the blocks have become fully replicated and then further blocks are added
to reach the limit defined in this parameter.
dfs.namenode.decommission.backoff.monitor.pending.blocks.per.lock
1000
When loading blocks into the replication queue, release the namenode write
lock after the defined number of blocks have been processed.
dfs.namenode.redundancy.interval.seconds
3
The periodicity in seconds with which the namenode computes
low redundancy work for datanodes. Support multiple time unit suffix(case insensitive),
as described in dfs.heartbeat.interval.
dfs.namenode.redundancy.queue.restart.iterations
2400
When picking blocks from the low redundancy queues, reset the
bookmarked iterator after the set number of iterations to ensure any blocks
which were not processed on the first pass are retried before the iterators
would naturally reach their end point. This ensures blocks are retried
more frequently when there are many pending blocks or blocks are
continuously added to the queues preventing the iterator reaching its
natural endpoint.
The default setting of 2400 combined with the default of
dfs.namenode.redundancy.interval.seconds means the iterators will be reset
approximately every 2 hours.
Setting this parameter to zero disables the feature and the iterators will
be reset only when the end of all queues has been reached.
dfs.namenode.accesstime.precision
3600000
The access time for HDFS file is precise upto this value.
The default value is 1 hour. Setting a value of 0 disables
access times for HDFS.
dfs.datanode.plugins
Comma-separated list of datanode plug-ins to be activated.
dfs.namenode.plugins
Comma-separated list of namenode plug-ins to be activated.
dfs.namenode.block-placement-policy.default.prefer-local-node
true
Controls how the default block placement policy places
the first replica of a block. When true, it will prefer the node where
the client is running. When false, it will prefer a node in the same rack
as the client. Setting to false avoids situations where entire copies of
large files end up on a single node, thus creating hotspots.
dfs.stream-buffer-size
4096
The size of buffer to stream files.
The size of this buffer should probably be a multiple of hardware
page size (4096 on Intel x86), and it determines how much data is
buffered during read and write operations.
dfs.bytes-per-checksum
512
The number of bytes per checksum. Must not be larger than
dfs.stream-buffer-size
dfs.client-write-packet-size
65536
Packet size for clients to write
dfs.client.write.exclude.nodes.cache.expiry.interval.millis
600000
The maximum period to keep a DN in the excluded nodes list
at a client. After this period, in milliseconds, the previously excluded node(s) will
be removed automatically from the cache and will be considered good for block allocations
again. Useful to lower or raise in situations where you keep a file open for very long
periods (such as a Write-Ahead-Log (WAL) file) to make the writer tolerant to cluster maintenance
restarts. Defaults to 10 minutes.
dfs.client.write.recover.lease.on.close.exception
false
Set to true to call recoverLease operation automatically when DFSOutputSteam closing encounters exception.
dfs.namenode.checkpoint.dir
file://${hadoop.tmp.dir}/dfs/namesecondary
Determines where on the local filesystem the DFS secondary
name node should store the temporary images to merge.
If this is a comma-delimited list of directories then the image is
replicated in all of the directories for redundancy.
dfs.namenode.checkpoint.edits.dir
${dfs.namenode.checkpoint.dir}
Determines where on the local filesystem the DFS secondary
name node should store the temporary edits to merge.
If this is a comma-delimited list of directories then the edits is
replicated in all of the directories for redundancy.
Default value is same as dfs.namenode.checkpoint.dir
dfs.namenode.checkpoint.period
3600
The number of seconds between two periodic checkpoints.
Support multiple time unit suffix(case insensitive), as described
in dfs.heartbeat.interval.If no time unit is specified then seconds
is assumed.
dfs.namenode.checkpoint.txns
1000000
The Secondary NameNode or CheckpointNode will create a checkpoint
of the namespace every 'dfs.namenode.checkpoint.txns' transactions, regardless
of whether 'dfs.namenode.checkpoint.period' has expired.
dfs.namenode.checkpoint.check.period
60
The SecondaryNameNode and CheckpointNode will poll the NameNode
every 'dfs.namenode.checkpoint.check.period' seconds to query the number
of uncheckpointed transactions. Support multiple time unit suffix(case insensitive),
as described in dfs.heartbeat.interval.If no time unit is specified then
seconds is assumed.
dfs.namenode.checkpoint.max-retries
3
The SecondaryNameNode retries failed checkpointing. If the
failure occurs while loading fsimage or replaying edits, the number of
retries is limited by this variable.
dfs.namenode.checkpoint.check.quiet-multiplier
1.5
Used to calculate the amount of time between retries when in the 'quiet' period
for creating checkpoints (active namenode already has an up-to-date image from another
checkpointer), so we wait a multiplier of the dfs.namenode.checkpoint.check.period before
retrying the checkpoint because another node likely is already managing the checkpoints,
allowing us to save bandwidth to transfer checkpoints that don't need to be used.
dfs.namenode.num.checkpoints.retained
2
The number of image checkpoint files (fsimage_*) that will be retained by
the NameNode and Secondary NameNode in their storage directories. All edit
logs (stored on edits_* files) necessary to recover an up-to-date namespace from the oldest retained
checkpoint will also be retained.
dfs.namenode.num.extra.edits.retained
1000000
The number of extra transactions which should be retained
beyond what is minimally necessary for a NN restart.
It does not translate directly to file's age, or the number of files kept,
but to the number of transactions (here "edits" means transactions).
One edit file may contain several transactions (edits).
During checkpoint, NameNode will identify the total number of edits to retain as extra by
checking the latest checkpoint transaction value, subtracted by the value of this property.
Then, it scans edits files to identify the older ones that don't include the computed range of
retained transactions that are to be kept around, and purges them subsequently.
The retainment can be useful for audit purposes or for an HA setup where a remote Standby Node may have
been offline for some time and need to have a longer backlog of retained
edits in order to start again.
Typically each edit is on the order of a few hundred bytes, so the default
of 1 million edits should be on the order of hundreds of MBs or low GBs.
NOTE: Fewer extra edits may be retained than value specified for this setting
if doing so would mean that more segments would be retained than the number
configured by dfs.namenode.max.extra.edits.segments.retained.
dfs.namenode.max.extra.edits.segments.retained
10000
The maximum number of extra edit log segments which should be retained
beyond what is minimally necessary for a NN restart. When used in conjunction with
dfs.namenode.num.extra.edits.retained, this configuration property serves to cap
the number of extra edits files to a reasonable value.
dfs.namenode.delegation.key.update-interval
86400000
The update interval for master key for delegation tokens
in the namenode in milliseconds.
dfs.namenode.delegation.token.max-lifetime
604800000
The maximum lifetime in milliseconds for which a delegation
token is valid.
dfs.namenode.delegation.token.renew-interval
86400000
The renewal interval for delegation token in milliseconds.
dfs.datanode.failed.volumes.tolerated
0
The number of volumes that are allowed to
fail before a datanode stops offering service. By default
any volume failure will cause a datanode to shutdown.
The value should be greater than or equal to -1 , -1 represents minimum
1 valid volume.
dfs.datanode.volumes.replica-add.threadpool.size
Specifies the maximum number of threads to use for
adding block in volume. Default value for this configuration is
max of (volume * number of bp_service, number of processor).
dfs.image.compress
false
When this value is true, the dfs image will be compressed.
Enabling this will be very helpful if dfs image is large since it can
avoid consuming a lot of network bandwidth when SBN uploads a new dfs
image to ANN. The compressed codec is specified by the setting
dfs.image.compression.codec.
dfs.image.compression.codec
org.apache.hadoop.io.compress.DefaultCodec
If the dfs image is compressed, how should they be compressed?
This has to be a codec defined in io.compression.codecs.
dfs.image.transfer.timeout
60000
Socket timeout for the HttpURLConnection instance used in the image
transfer. This is measured in milliseconds.
This timeout prevents client hangs if the connection is idle
for this configured timeout, during image transfer.
dfs.image.transfer.bandwidthPerSec
52428800
Maximum bandwidth used for regular image transfers (instead of
bootstrapping the standby namenode), in bytes per second.
This can help keep normal namenode operations responsive during
checkpointing.
A default value is 50mb per second.
The maximum bandwidth used for bootstrapping standby namenode is
configured with dfs.image.transfer-bootstrap-standby.bandwidthPerSec.
Support multiple size unit suffix(case insensitive), as described
in dfs.blocksize.
dfs.image.transfer-bootstrap-standby.bandwidthPerSec
0
Maximum bandwidth used for transferring image to bootstrap standby
namenode, in bytes per second.
A default value of 0 indicates that throttling is disabled. This default
value should be used in most cases, to ensure timely HA operations.
The maximum bandwidth used for regular image transfers is configured
with dfs.image.transfer.bandwidthPerSec.
Support multiple size unit suffix(case insensitive), as described in
dfs.blocksize.
dfs.image.transfer.chunksize
65536
Chunksize in bytes to upload the checkpoint.
Chunked streaming is used to avoid internal buffering of contents
of image file of huge size.
Support multiple size unit suffix(case insensitive), as described
in dfs.blocksize.
dfs.image.parallel.load
false
If true, write sub-section entries to the fsimage index so it can
be loaded in parallel. Also controls whether parallel loading
will be used for an image previously created with sub-sections.
If the image contains sub-sections and this is set to false,
parallel loading will not be used.
Parallel loading is not compatible with image compression,
so if dfs.image.compress is set to true this setting will be
ignored and no parallel loading will occur.
Enabling this feature may impact rolling upgrades and downgrades if
the previous version does not support this feature. If the feature was
enabled and a downgrade is required, first set this parameter to
false and then save the namespace to create a fsimage with no
sub-sections and then perform the downgrade.
dfs.image.parallel.target.sections
12
Controls the number of sub-sections that will be written to
fsimage for each section. This should be larger than
dfs.image.parallel.threads, otherwise all threads will not be
used when loading. Ideally, have at least twice the number
of target sections as threads, so each thread must load more
than one section to avoid one long running section affecting
the load time.
dfs.image.parallel.inode.threshold
1000000
If the image contains less inodes than this setting, then
do not write sub-sections and hence disable parallel loading.
This is because small images load very quickly in serial and
parallel loading is not needed.
dfs.image.parallel.threads
4
The number of threads to use when dfs.image.parallel.load is
enabled. This setting should be less than
dfs.image.parallel.target.sections. The optimal number of
threads will depend on the hardware and environment.
dfs.edit.log.transfer.timeout
30000
Socket timeout for edit log transfer in milliseconds. This timeout
should be configured such that normal edit log transfer for journal
node syncing can complete successfully.
dfs.edit.log.transfer.bandwidthPerSec
0
Maximum bandwidth used for transferring edit log to between journal nodes
for syncing, in bytes per second.
A default value of 0 indicates that throttling is disabled.
dfs.namenode.support.allow.format
true
Does HDFS namenode allow itself to be formatted?
You may consider setting this to false for any production
cluster, to avoid any possibility of formatting a running DFS.
dfs.datanode.max.transfer.threads
4096
Specifies the maximum number of threads to use for transferring data
in and out of the DN.
dfs.datanode.scan.period.hours
504
If this is positive, the DataNode will not scan any
individual block more than once in the specified scan period.
If this is negative, the block scanner is disabled.
If this is set to zero, then the default value of 504 hours
or 3 weeks is used. Prior versions of HDFS incorrectly documented
that setting this key to zero will disable the block scanner.
dfs.block.scanner.volume.bytes.per.second
1048576
If this is configured less than or equal to zero, the DataNode's block scanner will be disabled. If this
is positive, this is the number of bytes per second that the DataNode's
block scanner will try to scan from each volume.
dfs.block.scanner.skip.recent.accessed
false
If this is true, scanner will check the access time of block file to avoid
scanning blocks accessed during recent scan peroid, reducing disk IO.
This feature will not work if the DataNode volume has noatime mount option.
dfs.block.scanner.volume.join.timeout.ms
5000
The amount of time in milliseconds that the BlockScanner times out waiting
for the VolumeScanner thread to join during a shutdown call.
dfs.datanode.readahead.bytes
4194304
While reading block files, if the Hadoop native libraries are available,
the datanode can use the posix_fadvise system call to explicitly
page data into the operating system buffer cache ahead of the current
reader's position. This can improve performance especially when
disks are highly contended.
This configuration specifies the number of bytes ahead of the current
read position which the datanode will attempt to read ahead. This
feature may be disabled by configuring this property to 0.
If the native libraries are not available, this configuration has no
effect.
dfs.datanode.drop.cache.behind.reads
false
In some workloads, the data read from HDFS is known to be significantly
large enough that it is unlikely to be useful to cache it in the
operating system buffer cache. In this case, the DataNode may be
configured to automatically purge all data from the buffer cache
after it is delivered to the client. This behavior is automatically
disabled for workloads which read only short sections of a block
(e.g HBase random-IO workloads).
This may improve performance for some workloads by freeing buffer
cache space usage for more cacheable data.
If the Hadoop native libraries are not available, this configuration
has no effect.
dfs.datanode.drop.cache.behind.writes
false
In some workloads, the data written to HDFS is known to be significantly
large enough that it is unlikely to be useful to cache it in the
operating system buffer cache. In this case, the DataNode may be
configured to automatically purge all data from the buffer cache
after it is written to disk.
This may improve performance for some workloads by freeing buffer
cache space usage for more cacheable data.
If the Hadoop native libraries are not available, this configuration
has no effect.
dfs.datanode.sync.behind.writes
false
If this configuration is enabled, the datanode will instruct the
operating system to enqueue all written data to the disk immediately
after it is written. This differs from the usual OS policy which
may wait for up to 30 seconds before triggering writeback.
This may improve performance for some workloads by smoothing the
IO profile for data written to disk.
If the Hadoop native libraries are not available, this configuration
has no effect.
dfs.client.failover.max.attempts
15
Expert only. The number of client failover attempts that should be
made before the failover is considered failed.
dfs.client.failover.sleep.base.millis
500
Expert only. The time to wait, in milliseconds, between failover
attempts increases exponentially as a function of the number of
attempts made so far, with a random factor of +/- 50%. This option
specifies the base value used in the failover calculation. The
first failover will retry immediately. The 2nd failover attempt
will delay at least dfs.client.failover.sleep.base.millis
milliseconds. And so on.
dfs.client.failover.sleep.max.millis
15000
Expert only. The time to wait, in milliseconds, between failover
attempts increases exponentially as a function of the number of
attempts made so far, with a random factor of +/- 50%. This option
specifies the maximum value to wait between failovers.
Specifically, the time between two failover attempts will not
exceed +/- 50% of dfs.client.failover.sleep.max.millis
milliseconds.
dfs.client.failover.connection.retries
0
Expert only. Indicates the number of retries a failover IPC client
will make to establish a server connection.
dfs.client.failover.connection.retries.on.timeouts
0
Expert only. The number of retry attempts a failover IPC client
will make on socket timeout when establishing a server connection.
dfs.client.datanode-restart.timeout
30
Expert only. The time to wait, in seconds, from reception of an
datanode shutdown notification for quick restart, until declaring
the datanode dead and invoking the normal recovery mechanisms.
The notification is sent by a datanode when it is being shutdown
using the shutdownDatanode admin command with the upgrade option.
Support multiple time unit suffix(case insensitive), as described
in dfs.heartbeat.interval.If no time unit is specified then seconds
is assumed.
dfs.nameservices
Comma-separated list of nameservices.
dfs.nameservice.id
The ID of this nameservice. If the nameservice ID is not
configured or more than one nameservice is configured for
dfs.nameservices it is determined automatically by
matching the local node's address with the configured address.
dfs.internal.nameservices
Comma-separated list of nameservices that belong to this cluster.
Datanode will report to all the nameservices in this list. By default
this is set to the value of dfs.nameservices.
dfs.ha.namenodes.EXAMPLENAMESERVICE
The prefix for a given nameservice, contains a comma-separated
list of namenodes for a given nameservice (eg EXAMPLENAMESERVICE).
Unique identifiers for each NameNode in the nameservice, delimited by
commas. This will be used by DataNodes to determine all the NameNodes
in the cluster. For example, if you used “mycluster” as the nameservice
ID previously, and you wanted to use “nn1” and “nn2” as the individual
IDs of the NameNodes, you would configure a property
dfs.ha.namenodes.mycluster, and its value "nn1,nn2".
dfs.ha.namenode.id
The ID of this namenode. If the namenode ID is not configured it
is determined automatically by matching the local node's address
with the configured address.
dfs.ha.log-roll.period
120
How often, in seconds, the StandbyNode should ask the active to
roll edit logs. Since the StandbyNode only reads from finalized
log segments, the StandbyNode will only be as up-to-date as how
often the logs are rolled. Note that failover triggers a log roll
so the StandbyNode will be up to date before it becomes active.
Support multiple time unit suffix(case insensitive), as described
in dfs.heartbeat.interval.If no time unit is specified then seconds
is assumed.
dfs.ha.tail-edits.period
60
How often, the StandbyNode and ObserverNode should check if there are new
edit log entries ready to be consumed. This is the minimum period between
checking; exponential backoff will be applied if no edits are found and
dfs.ha.tail-edits.period.backoff-max is configured. By default, no
backoff is applied.
Supports multiple time unit suffix (case insensitive), as described
in dfs.heartbeat.interval.
dfs.ha.tail-edits.period.backoff-max
0
The maximum time the tailer should wait between checking for new edit log
entries. Exponential backoff will be applied when an edit log tail is
performed but no edits are available to be read. Values less than or
equal to zero disable backoff entirely; this is the default behavior.
Supports multiple time unit suffix (case insensitive), as described
in dfs.heartbeat.interval.
dfs.ha.tail-edits.namenode-retries
3
Number of retries to use when contacting the namenode when tailing the log.
dfs.ha.tail-edits.rolledits.timeout
60
The timeout in seconds of calling rollEdits RPC on Active NN.
dfs.ha.automatic-failover.enabled
false
Whether automatic failover is enabled. See the HDFS High
Availability documentation for details on automatic HA
configuration.
dfs.client.use.datanode.hostname
false
Whether clients should use datanode hostnames when
connecting to datanodes.
dfs.datanode.use.datanode.hostname
false
Whether datanodes should use datanode hostnames when
connecting to other datanodes for data transfer.
dfs.client.local.interfaces
A comma separated list of network interface names to use
for data transfer between the client and datanodes. When creating
a connection to read from or write to a datanode, the client
chooses one of the specified interfaces at random and binds its
socket to the IP of that interface. Individual names may be
specified as either an interface name (eg "eth0"), a subinterface
name (eg "eth0:0"), or an IP address (which may be specified using
CIDR notation to match a range of IPs).
dfs.datanode.shared.file.descriptor.paths
/dev/shm,/tmp
A comma-separated list of paths to use when creating file descriptors that
will be shared between the DataNode and the DFSClient. Typically we use
/dev/shm, so that the file descriptors will not be written to disk.
It tries paths in order until creation of shared memory segment succeeds.
dfs.short.circuit.shared.memory.watcher.interrupt.check.ms
60000
The length of time in milliseconds that the short-circuit shared memory
watcher will go between checking for java interruptions sent from other
threads. This is provided mainly for unit tests.
dfs.namenode.kerberos.principal
The NameNode service principal. This is typically set to
nn/_HOST@REALM.TLD. Each NameNode will substitute _HOST with its
own fully qualified hostname at startup. The _HOST placeholder
allows using the same configuration setting on both NameNodes
in an HA setup.
dfs.namenode.keytab.file
The keytab file used by each NameNode daemon to login as its
service principal. The principal name is configured with
dfs.namenode.kerberos.principal.
dfs.datanode.kerberos.principal
The DataNode service principal. This is typically set to
dn/_HOST@REALM.TLD. Each DataNode will substitute _HOST with its
own fully qualified hostname at startup. The _HOST placeholder
allows using the same configuration setting on all DataNodes.
dfs.datanode.keytab.file
The keytab file used by each DataNode daemon to login as its
service principal. The principal name is configured with
dfs.datanode.kerberos.principal.
dfs.journalnode.kerberos.principal
The JournalNode service principal. This is typically set to
jn/_HOST@REALM.TLD. Each JournalNode will substitute _HOST with its
own fully qualified hostname at startup. The _HOST placeholder
allows using the same configuration setting on all JournalNodes.
dfs.journalnode.keytab.file
The keytab file used by each JournalNode daemon to login as its
service principal. The principal name is configured with
dfs.journalnode.kerberos.principal.
dfs.namenode.kerberos.internal.spnego.principal
${dfs.web.authentication.kerberos.principal}
The server principal used by the NameNode for web UI SPNEGO
authentication when Kerberos security is enabled. This is
typically set to HTTP/_HOST@REALM.TLD The SPNEGO server principal
begins with the prefix HTTP/ by convention.
If the value is '*', the web server will attempt to login with
every principal specified in the keytab file
dfs.web.authentication.kerberos.keytab.
dfs.journalnode.kerberos.internal.spnego.principal
The server principal used by the JournalNode HTTP Server for
SPNEGO authentication when Kerberos security is enabled. This is
typically set to HTTP/_HOST@REALM.TLD. The SPNEGO server principal
begins with the prefix HTTP/ by convention.
If the value is '*', the web server will attempt to login with
every principal specified in the keytab file
dfs.web.authentication.kerberos.keytab.
For most deployments this can be set to ${dfs.web.authentication.kerberos.principal}
i.e use the value of dfs.web.authentication.kerberos.principal.
dfs.secondary.namenode.kerberos.internal.spnego.principal
${dfs.web.authentication.kerberos.principal}
The server principal used by the Secondary NameNode for web UI SPNEGO
authentication when Kerberos security is enabled. Like all other
Secondary NameNode settings, it is ignored in an HA setup.
If the value is '*', the web server will attempt to login with
every principal specified in the keytab file
dfs.web.authentication.kerberos.keytab.
dfs.web.authentication.kerberos.principal
The server principal used by the NameNode for WebHDFS SPNEGO
authentication.
Required when WebHDFS and security are enabled. In most secure clusters this
setting is also used to specify the values for
dfs.namenode.kerberos.internal.spnego.principal and
dfs.journalnode.kerberos.internal.spnego.principal.
dfs.web.authentication.kerberos.keytab
The keytab file for the principal corresponding to
dfs.web.authentication.kerberos.principal.
dfs.namenode.kerberos.principal.pattern
*
A client-side RegEx that can be configured to control
allowed realms to authenticate with (useful in cross-realm env.)
dfs.namenode.avoid.read.stale.datanode
false
Indicate whether or not to avoid reading from "stale" datanodes whose
heartbeat messages have not been received by the namenode
for more than a specified time interval. Stale datanodes will be
moved to the end of the node list returned for reading. See
dfs.namenode.avoid.write.stale.datanode for a similar setting for writes.
dfs.namenode.avoid.read.slow.datanode
false
Indicate whether or not to avoid reading from "slow" datanodes.
Slow datanodes will be moved to the end of the node list returned
for reading.
dfs.namenode.avoid.write.stale.datanode
false
Indicate whether or not to avoid writing to "stale" datanodes whose
heartbeat messages have not been received by the namenode
for more than a specified time interval. Writes will avoid using
stale datanodes unless more than a configured ratio
(dfs.namenode.write.stale.datanode.ratio) of datanodes are marked as
stale. See dfs.namenode.avoid.read.stale.datanode for a similar setting
for reads.
dfs.namenode.enable.log.stale.datanode
false
Enable and disable logging datanode staleness. Disabled by default.
dfs.namenode.stale.datanode.interval
30000
Default time interval in milliseconds for marking a datanode as "stale",
i.e., if the namenode has not received heartbeat msg from a datanode for
more than this time interval, the datanode will be marked and treated
as "stale" by default. The stale interval cannot be too small since
otherwise this may cause too frequent change of stale states.
We thus set a minimum stale interval value (the default value is 3 times
of heartbeat interval) and guarantee that the stale interval cannot be less
than the minimum value. A stale data node is avoided during lease/block
recovery. It can be conditionally avoided for reads (see
dfs.namenode.avoid.read.stale.datanode) and for writes (see
dfs.namenode.avoid.write.stale.datanode).
dfs.namenode.write.stale.datanode.ratio
0.5f
When the ratio of number stale datanodes to total datanodes marked
is greater than this ratio, stop avoiding writing to stale nodes so
as to prevent causing hotspots.
dfs.namenode.invalidate.work.pct.per.iteration
0.32f
*Note*: Advanced property. Change with caution.
This determines the percentage amount of block
invalidations (deletes) to do over a single DN heartbeat
deletion command. The final deletion count is determined by applying this
percentage to the number of live nodes in the system.
The resultant number is the number of blocks from the deletion list
chosen for proper invalidation over a single heartbeat of a single DN.
Value should be a positive, non-zero percentage in float notation (X.Yf),
with 1.0f meaning 100%.
dfs.namenode.replication.work.multiplier.per.iteration
2
*Note*: Advanced property. Change with caution.
This determines the total amount of block transfers to begin in
parallel at a DN, for replication, when such a command list is being
sent over a DN heartbeat by the NN. The actual number is obtained by
multiplying this multiplier with the total number of live nodes in the
cluster. The result number is the number of blocks to begin transfers
immediately for, per DN heartbeat. This number can be any positive,
non-zero integer.
nfs.server.port
2049
Specify the port number used by Hadoop NFS.
nfs.mountd.port
4242
Specify the port number used by Hadoop mount daemon.
nfs.dump.dir
/tmp/.hdfs-nfs
This directory is used to temporarily save out-of-order writes before
writing to HDFS. For each file, the out-of-order writes are dumped after
they are accumulated to exceed certain threshold (e.g., 1MB) in memory.
One needs to make sure the directory has enough space.
nfs.rtmax
1048576
This is the maximum size in bytes of a READ request
supported by the NFS gateway. If you change this, make sure you
also update the nfs mount's rsize(add rsize= # of bytes to the
mount directive).
nfs.wtmax
1048576
This is the maximum size in bytes of a WRITE request
supported by the NFS gateway. If you change this, make sure you
also update the nfs mount's wsize(add wsize= # of bytes to the
mount directive).
nfs.keytab.file
*Note*: Advanced property. Change with caution.
This is the path to the keytab file for the hdfs-nfs gateway.
This is required when the cluster is kerberized.
nfs.kerberos.principal
*Note*: Advanced property. Change with caution.
This is the name of the kerberos principal. This is required when
the cluster is kerberized.It must be of this format:
nfs-gateway-user/nfs-gateway-host@kerberos-realm
nfs.allow.insecure.ports
true
When set to false, client connections originating from unprivileged ports
(those above 1023) will be rejected. This is to ensure that clients
connecting to this NFS Gateway must have had root privilege on the machine
where they're connecting from.
hadoop.fuse.connection.timeout
300
The minimum number of seconds that we'll cache libhdfs connection objects
in fuse_dfs. Lower values will result in lower memory consumption; higher
values may speed up access by avoiding the overhead of creating new
connection objects.
hadoop.fuse.timer.period
5
The number of seconds between cache expiry checks in fuse_dfs. Lower values
will result in fuse_dfs noticing changes to Kerberos ticket caches more
quickly.
dfs.namenode.metrics.logger.period.seconds
600
This setting controls how frequently the NameNode logs its metrics. The
logging configuration must also define one or more appenders for
NameNodeMetricsLog for the metrics to be logged.
NameNode metrics logging is disabled if this value is set to zero or
less than zero.
dfs.datanode.metrics.logger.period.seconds
600
This setting controls how frequently the DataNode logs its metrics. The
logging configuration must also define one or more appenders for
DataNodeMetricsLog for the metrics to be logged.
DataNode metrics logging is disabled if this value is set to zero or
less than zero.
dfs.metrics.percentiles.intervals
Comma-delimited set of integers denoting the desired rollover intervals
(in seconds) for percentile latency metrics on the Namenode and Datanode.
By default, percentile latency metrics are disabled.
dfs.datanode.peer.stats.enabled
false
A switch to turn on/off tracking DataNode peer statistics.
dfs.datanode.peer.metrics.min.outlier.detection.samples
1000
Minimum number of packet send samples which are required to qualify for outlier detection.
If the number of samples is below this then outlier detection is skipped.
dfs.datanode.min.outlier.detection.nodes
10
Minimum number of nodes to run outlier detection.
dfs.datanode.slowpeer.low.threshold.ms
5
Threshold in milliseconds below which a DataNode is definitely not slow.
dfs.datanode.max.nodes.to.report
5
Number of nodes to include in JSON report. We will return nodes with
the highest number of votes from peers.
dfs.datanode.outliers.report.interval
30m
This setting controls how frequently DataNodes will report their peer
latencies to the NameNode via heartbeats. This setting supports
multiple time unit suffixes as described in dfs.heartbeat.interval.
If no suffix is specified then milliseconds is assumed.
It is ignored if dfs.datanode.peer.stats.enabled is false.
dfs.namenode.block-placement-policy.exclude-slow-nodes.enabled
false
If this is set to true, we will filter out slow nodes
when choosing targets for blocks.
dfs.namenode.max.slowpeer.collect.nodes
5
How many slow nodes we will collect for filtering out
when choosing targets for blocks.
It is ignored if dfs.namenode.block-placement-policy.exclude-slow-nodes.enabled is false.
dfs.namenode.slowpeer.collect.interval
30m
Interval at which the slow peer trackers runs in the background to collect slow peers.
It is ignored if dfs.namenode.block-placement-policy.exclude-slow-nodes.enabled is false.
dfs.datanode.fileio.profiling.sampling.percentage
0
This setting controls the percentage of file I/O events which will be
profiled for DataNode disk statistics. The default value of 0 disables
disk statistics. Set to an integer value between 1 and 100 to enable disk
statistics.
dfs.datanode.min.outlier.detection.disks
5
Minimum number of disks to run outlier detection.
dfs.datanode.slowdisk.low.threshold.ms
20
Threshold in milliseconds below which a disk is definitely not slow.
dfs.datanode.max.disks.to.report
5
Number of disks to include in JSON report per operation. We will return
disks with the highest latency.
dfs.datanode.max.slowdisks.to.exclude
0
The number of slow disks that needs to be excluded. By default, this parameter is set to 0,
which disables excluding slow disk when choosing volume.
hadoop.user.group.metrics.percentiles.intervals
A comma-separated list of the granularity in seconds for the metrics
which describe the 50/75/90/95/99th percentile latency for group resolution
in milliseconds.
By default, percentile latency metrics are disabled.
dfs.encrypt.data.transfer
false
Whether or not actual block data that is read/written from/to HDFS should
be encrypted on the wire. This only needs to be set on the NN and DNs,
clients will deduce this automatically. It is possible to override this setting
per connection by specifying custom logic via dfs.trustedchannel.resolver.class.
dfs.encrypt.data.transfer.algorithm
This value may be set to either "3des" or "rc4". If nothing is set, then
the configured JCE default on the system is used (usually 3DES.) It is
widely believed that 3DES is more cryptographically secure, but RC4 is
substantially faster.
Note that if AES is supported by both the client and server then this
encryption algorithm will only be used to initially transfer keys for AES.
(See dfs.encrypt.data.transfer.cipher.suites.)
dfs.encrypt.data.transfer.cipher.suites
This value may be either undefined or AES/CTR/NoPadding. If defined, then
dfs.encrypt.data.transfer uses the specified cipher suite for data
encryption. If not defined, then only the algorithm specified in
dfs.encrypt.data.transfer.algorithm is used. By default, the property is
not defined.
dfs.encrypt.data.transfer.cipher.key.bitlength
128
The key bitlength negotiated by dfsclient and datanode for encryption.
This value may be set to either 128, 192 or 256.
dfs.trustedchannel.resolver.class
TrustedChannelResolver is used to determine whether a channel
is trusted for plain data transfer. The TrustedChannelResolver is
invoked on both client and server side. If the resolver indicates
that the channel is trusted, then the data transfer will not be
encrypted even if dfs.encrypt.data.transfer is set to true. The
default implementation returns false indicating that the channel
is not trusted.
dfs.data.transfer.protection
A comma-separated list of SASL protection values used for secured
connections to the DataNode when reading or writing block data. Possible
values are authentication, integrity and privacy. authentication means
authentication only and no integrity or privacy; integrity implies
authentication and integrity are enabled; and privacy implies all of
authentication, integrity and privacy are enabled. If
dfs.encrypt.data.transfer is set to true, then it supersedes the setting for
dfs.data.transfer.protection and enforces that all connections must use a
specialized encrypted SASL handshake. This property is ignored for
connections to a DataNode listening on a privileged port. In this case, it
is assumed that the use of a privileged port establishes sufficient trust.
dfs.data.transfer.saslproperties.resolver.class
SaslPropertiesResolver used to resolve the QOP used for a connection to the
DataNode when reading or writing block data. If not specified, the value of
hadoop.security.saslproperties.resolver.class is used as the default value.
dfs.journalnode.rpc-address
0.0.0.0:8485
The JournalNode RPC server address and port.
dfs.journalnode.rpc-bind-host
The actual address the RPC server will bind to. If this optional address is
set, it overrides only the hostname portion of dfs.journalnode.rpc-address.
This is useful for making the JournalNode listen on all interfaces by
setting it to 0.0.0.0.
dfs.journalnode.http-address
0.0.0.0:8480
The address and port the JournalNode HTTP server listens on.
If the port is 0 then the server will start on a free port.
dfs.journalnode.http-bind-host
The actual address the HTTP server will bind to. If this optional address
is set, it overrides only the hostname portion of
dfs.journalnode.http-address. This is useful for making the JournalNode
HTTP server listen on allinterfaces by setting it to 0.0.0.0.
dfs.journalnode.https-address
0.0.0.0:8481
The address and port the JournalNode HTTPS server listens on.
If the port is 0 then the server will start on a free port.
dfs.journalnode.https-bind-host
The actual address the HTTP server will bind to. If this optional address
is set, it overrides only the hostname portion of
dfs.journalnode.https-address. This is useful for making the JournalNode
HTTP server listen on all interfaces by setting it to 0.0.0.0.
dfs.namenode.audit.loggers
default
List of classes implementing audit loggers that will receive audit events.
These should be implementations of org.apache.hadoop.hdfs.server.namenode.AuditLogger.
The special value "default" can be used to reference the default audit
logger, which uses the configured log system. Installing custom audit loggers
may affect the performance and stability of the NameNode. Refer to the custom
logger's documentation for more details.
dfs.datanode.available-space-volume-choosing-policy.balanced-space-threshold
10737418240
Only used when the dfs.datanode.fsdataset.volume.choosing.policy is set to
org.apache.hadoop.hdfs.server.datanode.fsdataset.AvailableSpaceVolumeChoosingPolicy.
This setting controls how much DN volumes are allowed to differ in terms of
bytes of free disk space before they are considered imbalanced. If the free
space of all the volumes are within this range of each other, the volumes
will be considered balanced and block assignments will be done on a pure
round robin basis. Support multiple size unit suffix(case insensitive), as
described in dfs.blocksize.
dfs.datanode.available-space-volume-choosing-policy.balanced-space-preference-fraction
0.75f
Only used when the dfs.datanode.fsdataset.volume.choosing.policy is set to
org.apache.hadoop.hdfs.server.datanode.fsdataset.AvailableSpaceVolumeChoosingPolicy.
This setting controls what percentage of new block allocations will be sent
to volumes with more available disk space than others. This setting should
be in the range 0.0 - 1.0, though in practice 0.5 - 1.0, since there should
be no reason to prefer that volumes with less available disk space receive
more block allocations.
dfs.datanode.round-robin-volume-choosing-policy.additional-available-space
1073741824
Only used when the dfs.datanode.fsdataset.volume.choosing.policy is set to
org.apache.hadoop.hdfs.server.datanode.fsdataset.RoundRobinVolumeChoosingPolicy.
This setting controls how much additional available space (unit is byte) is needed
when choosing a volume.
dfs.namenode.edits.noeditlogchannelflush
false
Specifies whether to flush edit log file channel. When set, expensive
FileChannel#force calls are skipped and synchronous disk writes are
enabled instead by opening the edit log file with RandomAccessFile("rwd")
flags. This can significantly improve the performance of edit log writes
on the Windows platform.
Note that the behavior of the "rwd" flags is platform and hardware specific
and might not provide the same level of guarantees as FileChannel#force.
For example, the write will skip the disk-cache on SAS and SCSI devices
while it might not on SATA devices. This is an expert level setting,
change with caution.
dfs.client.cache.drop.behind.writes
Just like dfs.datanode.drop.cache.behind.writes, this setting causes the
page cache to be dropped behind HDFS writes, potentially freeing up more
memory for other uses. Unlike dfs.datanode.drop.cache.behind.writes, this
is a client-side setting rather than a setting for the entire datanode.
If present, this setting will override the DataNode default.
If the native libraries are not available to the DataNode, this
configuration has no effect.
dfs.client.cache.drop.behind.reads
Just like dfs.datanode.drop.cache.behind.reads, this setting causes the
page cache to be dropped behind HDFS reads, potentially freeing up more
memory for other uses. Unlike dfs.datanode.drop.cache.behind.reads, this
is a client-side setting rather than a setting for the entire datanode. If
present, this setting will override the DataNode default.
If the native libraries are not available to the DataNode, this
configuration has no effect.
dfs.client.cache.readahead
When using remote reads, this setting causes the datanode to
read ahead in the block file using posix_fadvise, potentially decreasing
I/O wait times. Unlike dfs.datanode.readahead.bytes, this is a client-side
setting rather than a setting for the entire datanode. If present, this
setting will override the DataNode default. Support multiple size unit
suffix(case insensitive), as described in dfs.blocksize.
When using local reads, this setting determines how much readahead we do in
BlockReaderLocal.
If the native libraries are not available to the DataNode, this
configuration has no effect.
dfs.client.server-defaults.validity.period.ms
3600000
The amount of milliseconds after which cached server defaults are updated.
By default this parameter is set to 1 hour.
Support multiple time unit suffix(case insensitive), as described
in dfs.heartbeat.interval.
dfs.namenode.enable.retrycache
true
This enables the retry cache on the namenode. Namenode tracks for
non-idempotent requests the corresponding response. If a client retries the
request, the response from the retry cache is sent. Such operations
are tagged with annotation @AtMostOnce in namenode protocols. It is
recommended that this flag be set to true. Setting it to false, will result
in clients getting failure responses to retried request. This flag must
be enabled in HA setup for transparent fail-overs.
The entries in the cache have expiration time configurable
using dfs.namenode.retrycache.expirytime.millis.
dfs.namenode.retrycache.expirytime.millis
600000
The time for which retry cache entries are retained.
dfs.namenode.retrycache.heap.percent
0.03f
This parameter configures the heap size allocated for retry cache
(excluding the response cached). This corresponds to approximately
4096 entries for every 64MB of namenode process java heap size.
Assuming retry cache entry expiration time (configured using
dfs.namenode.retrycache.expirytime.millis) of 10 minutes, this
enables retry cache to support 7 operations per second sustained
for 10 minutes. As the heap size is increased, the operation rate
linearly increases.
dfs.client.mmap.enabled
true
If this is set to false, the client won't attempt to perform memory-mapped reads.
dfs.client.mmap.cache.size
256
When zero-copy reads are used, the DFSClient keeps a cache of recently used
memory mapped regions. This parameter controls the maximum number of
entries that we will keep in that cache.
The larger this number is, the more file descriptors we will potentially
use for memory-mapped files. mmaped files also use virtual address space.
You may need to increase your ulimit virtual address space limits before
increasing the client mmap cache size.
Note that you can still do zero-copy reads when this size is set to 0.
dfs.client.mmap.cache.timeout.ms
3600000
The minimum length of time that we will keep an mmap entry in the cache
between uses. If an entry is in the cache longer than this, and nobody
uses it, it will be removed by a background thread.
dfs.client.mmap.retry.timeout.ms
300000
The minimum amount of time that we will wait before retrying a failed mmap
operation.
dfs.client.short.circuit.replica.stale.threshold.ms
1800000
The maximum amount of time that we will consider a short-circuit replica to
be valid, if there is no communication from the DataNode. After this time
has elapsed, we will re-fetch the short-circuit replica even if it is in
the cache.
dfs.namenode.caching.enabled
true
Set to true to enable block caching. This flag enables the NameNode to
maintain a mapping of cached blocks to DataNodes via processing DataNode
cache reports. Based on these reports and addition and removal of caching
directives, the NameNode will schedule caching and uncaching work.
In the current implementation, centralized caching introduces additional
write lock overhead (see CacheReplicationMonitor#rescan) even if no path
to cache is specified, so we recommend disabling this feature when not in
use. We will disable centralized caching by default in later versions.
dfs.namenode.path.based.cache.block.map.allocation.percent
0.25
The percentage of the Java heap which we will allocate to the cached blocks
map. The cached blocks map is a hash map which uses chained hashing.
Smaller maps may be accessed more slowly if the number of cached blocks is
large; larger maps will consume more memory.
dfs.datanode.max.locked.memory
0
The amount of memory in bytes to use for caching of block replicas in
memory on the datanode. The datanode's maximum locked memory soft ulimit
(RLIMIT_MEMLOCK) must be set to at least this value, else the datanode
will abort on startup. Support multiple size unit suffix(case insensitive),
as described in dfs.blocksize.
By default, this parameter is set to 0, which disables in-memory caching.
If the native libraries are not available to the DataNode, this
configuration has no effect.
dfs.datanode.pmem.cache.dirs
This value specifies the persistent memory directory used for caching block
replica. Multiple directories separated by "," are acceptable.
dfs.datanode.pmem.cache.recovery
true
This value specifies whether previous cache on persistent memory will be recovered.
This configuration can take effect only if persistent memory cache is enabled by
specifying value for 'dfs.datanode.pmem.cache.dirs'.
dfs.namenode.list.cache.directives.num.responses
100
This value controls the number of cache directives that the NameNode will
send over the wire in response to a listDirectives RPC.
dfs.namenode.list.cache.pools.num.responses
100
This value controls the number of cache pools that the NameNode will
send over the wire in response to a listPools RPC.
dfs.namenode.path.based.cache.refresh.interval.ms
30000
The amount of milliseconds between subsequent path cache rescans. Path
cache rescans are when we calculate which blocks should be cached, and on
what datanodes.
By default, this parameter is set to 30 seconds.
dfs.namenode.path.based.cache.retry.interval.ms
30000
When the NameNode needs to uncache something that is cached, or cache
something that is not cached, it must direct the DataNodes to do so by
sending a DNA_CACHE or DNA_UNCACHE command in response to a DataNode
heartbeat. This parameter controls how frequently the NameNode will
resend these commands.
dfs.datanode.fsdatasetcache.max.threads.per.volume
4
The maximum number of threads per volume to use for caching new data
on the datanode. These threads consume both I/O and CPU. This can affect
normal datanode operations.
dfs.datanode.fsdatasetasyncdisk.max.threads.per.volume
4
The maximum number of threads per volume used to process async disk
operations on the datanode. These threads consume I/O and CPU at the
same time. This will affect normal data node operations.
dfs.cachereport.intervalMsec
10000
Determines cache reporting interval in milliseconds. After this amount of
time, the DataNode sends a full report of its cache state to the NameNode.
The NameNode uses the cache report to update its map of cached blocks to
DataNode locations.
This configuration has no effect if in-memory caching has been disabled by
setting dfs.datanode.max.locked.memory to 0 (which is the default).
If the native libraries are not available to the DataNode, this
configuration has no effect.
dfs.namenode.edit.log.autoroll.multiplier.threshold
0.5
Determines when an active namenode will roll its own edit log.
The actual threshold (in number of edits) is determined by multiplying
this value by dfs.namenode.checkpoint.txns.
This prevents extremely large edit files from accumulating on the active
namenode, which can cause timeouts during namenode startup and pose an
administrative hassle. This behavior is intended as a failsafe for when
the standby or secondary namenode fail to roll the edit log by the normal
checkpoint threshold.
dfs.namenode.edit.log.autoroll.check.interval.ms
300000
How often an active namenode will check if it needs to roll its edit log,
in milliseconds.
dfs.webhdfs.user.provider.user.pattern
^[A-Za-z_][A-Za-z0-9._-]*[$]?$
Valid pattern for user and group names for webhdfs, it must be a valid java regex.
dfs.webhdfs.acl.provider.permission.pattern
^(default:)?(user|group|mask|other):[[A-Za-z_][A-Za-z0-9._-]]*:([rwx-]{3})?(,(default:)?(user|group|mask|other):[[A-Za-z_][A-Za-z0-9._-]]*:([rwx-]{3})?)*$
Valid pattern for user and group names in webhdfs acl operations, it must be a valid java regex.
dfs.webhdfs.socket.connect-timeout
60s
Socket timeout for connecting to WebHDFS servers. This prevents a
WebHDFS client from hanging if the server hostname is
misconfigured, or the server does not response before the timeout
expires. Value is followed by a unit specifier: ns, us, ms, s, m,
h, d for nanoseconds, microseconds, milliseconds, seconds,
minutes, hours, days respectively. Values should provide units,
but milliseconds are assumed.
dfs.webhdfs.socket.read-timeout
60s
Socket timeout for reading data from WebHDFS servers. This
prevents a WebHDFS client from hanging if the server stops sending
data. Value is followed by a unit specifier: ns, us, ms, s, m, h,
d for nanoseconds, microseconds, milliseconds, seconds, minutes,
hours, days respectively. Values should provide units,
but milliseconds are assumed.
dfs.client.context
default
The name of the DFSClient context that we should use. Clients that share
a context share a socket cache and short-circuit cache, among other things.
You should only change this if you don't want to share with another set of
threads.
dfs.client.read.shortcircuit
false
This configuration parameter turns on short-circuit local reads.
dfs.client.socket.send.buffer.size
0
Socket send buffer size for a write pipeline in DFSClient side.
This may affect TCP connection throughput.
If it is set to zero or negative value,
no buffer size will be set explicitly,
thus enable tcp auto-tuning on some system.
The default value is 0.
dfs.domain.socket.path
Optional. This is a path to a UNIX domain socket that will be used for
communication between the DataNode and local HDFS clients.
If the string "_PORT" is present in this path, it will be replaced by the
TCP port of the DataNode.
dfs.domain.socket.disable.interval.seconds
600
The interval that a DataNode is disabled for future Short-Circuit Reads,
after an error happens during a Short-Circuit Read. Setting this to 0 will
not disable Short-Circuit Reads at all after errors happen. Negative values
are invalid.
dfs.client.read.shortcircuit.skip.checksum
false
If this configuration parameter is set,
short-circuit local reads will skip checksums.
This is normally not recommended,
but it may be useful for special setups.
You might consider using this
if you are doing your own checksumming outside of HDFS.
dfs.client.read.shortcircuit.streams.cache.size
256
The DFSClient maintains a cache of recently opened file descriptors.
This parameter controls the maximum number of file descriptors in the cache.
Setting this higher will use more file descriptors,
but potentially provide better performance on workloads
involving lots of seeks.
dfs.client.read.shortcircuit.streams.cache.expiry.ms
300000
This controls the minimum amount of time
file descriptors need to sit in the client cache context
before they can be closed for being inactive for too long.
dfs.namenode.audit.log.debug.cmdlist
A comma separated list of NameNode commands that are written to the HDFS
namenode audit log only if the audit log level is debug.
dfs.client.use.legacy.blockreader.local
false
Legacy short-circuit reader implementation based on HDFS-2246 is used
if this configuration parameter is true.
This is for the platforms other than Linux
where the new implementation based on HDFS-347 is not available.
dfs.client.read.use.cache.priority
false
If true, the cached replica of the datanode is preferred
else the replica closest to client is preferred.
dfs.block.local-path-access.user
Comma separated list of the users allowed to open block files
on legacy short-circuit local read.
dfs.client.domain.socket.data.traffic
false
This control whether we will try to pass normal data traffic
over UNIX domain socket rather than over TCP socket
on node-local data transfer.
This is currently experimental and turned off by default.
dfs.namenode.reject-unresolved-dn-topology-mapping
false
If the value is set to true, then namenode will reject datanode
registration if the topology mapping for a datanode is not resolved and
NULL is returned (script defined by net.topology.script.file.name fails
to execute). Otherwise, datanode will be registered and the default rack
will be assigned as the topology path. Topology paths are important for
data resiliency, since they define fault domains. Thus it may be unwanted
behavior to allow datanode registration with the default rack if the
resolving topology failed.
dfs.namenode.xattrs.enabled
true
Whether support for extended attributes is enabled on the NameNode.
dfs.namenode.fs-limits.max-xattrs-per-inode
32
Maximum number of extended attributes per inode.
dfs.namenode.fs-limits.max-xattr-size
16384
The maximum combined size of the name and value of an extended attribute
in bytes. It should be larger than 0, and less than or equal to maximum
size hard limit which is 32768.
Support multiple size unit suffix(case insensitive), as described in
dfs.blocksize.
dfs.client.slow.io.warning.threshold.ms
30000
The threshold in milliseconds at which we will log a slow
io warning in a dfsclient. By default, this parameter is set to 30000
milliseconds (30 seconds).
dfs.datanode.slow.io.warning.threshold.ms
300
The threshold in milliseconds at which we will log a slow
io warning in a datanode. By default, this parameter is set to 300
milliseconds.
dfs.datanode.processcommands.threshold
2s
The threshold in milliseconds at which we will log a slow
command processing in BPServiceActor. By default, this parameter is set
to 2 seconds.
dfs.client.deadnode.detection.enabled
false
Set to true to enable dead node detection in client side. Then all the DFSInputStreams of the same client can
share the dead node information.
dfs.client.deadnode.detection.probe.deadnode.threads
10
The maximum number of threads to use for probing dead node.
dfs.client.deadnode.detection.idle.sleep.ms
10000
The sleep time of DeadNodeDetector per iteration.
dfs.client.deadnode.detection.probe.suspectnode.threads
10
The maximum number of threads to use for probing suspect node.
dfs.client.deadnode.detection.rpc.threads
20
The maximum number of threads to use for calling RPC call to recheck the liveness of dead node.
dfs.client.deadnode.detection.probe.deadnode.interval.ms
60000
Interval time in milliseconds for probing dead node behavior.
dfs.client.deadnode.detection.probe.suspectnode.interval.ms
300
Interval time in milliseconds for probing suspect node behavior.
dfs.client.deadnode.detection.probe.connection.timeout.ms
20000
Connection timeout for probing dead node in milliseconds.
dfs.client.refresh.read-block-locations.ms
0
Refreshing LocatedBlocks period. A value of 0 disables the feature.
dfs.client.refresh.read-block-locations.register-automatically
true
Whether to auto-register all DFSInputStreams for background LocatedBlock refreshes.
If false, user must manually register using DFSClient#addLocatedBlocksRefresh(DFSInputStream)
dfs.client.refresh.read-block-locations.threads
5
Number of threads to use for refreshing LocatedBlocks of registered
DFSInputStreams. If a DFSClient opens many DFSInputStreams, increasing
this may help refresh them all in a timely manner.
dfs.namenode.lease-recheck-interval-ms
2000
During the release of lease a lock is hold that make any
operations on the namenode stuck. In order to not block them during
a too long duration we stop releasing lease after this max lock limit.
dfs.namenode.max-lock-hold-to-release-lease-ms
25
During the release of lease a lock is hold that make any
operations on the namenode stuck. In order to not block them during
a too long duration we stop releasing lease after this max lock limit.
dfs.namenode.write-lock-reporting-threshold-ms
5000
When a write lock is held on the namenode for a long time,
this will be logged as the lock is released. This sets how long the
lock must be held for logging to occur.
dfs.namenode.read-lock-reporting-threshold-ms
5000
When a read lock is held on the namenode for a long time,
this will be logged as the lock is released. This sets how long the
lock must be held for logging to occur.
dfs.namenode.access-control-enforcer-reporting-threshold-ms
1000
If an external AccessControlEnforcer runs for a long time to check permission with the FSnamesystem lock,
print a WARN log message. This sets how long must be run for logging to occur.
dfs.namenode.lock.detailed-metrics.enabled
false
If true, the namenode will keep track of how long various
operations hold the Namesystem lock for and emit this as metrics. These
metrics have names of the form FSN(Read|Write)LockNanosOperationName,
where OperationName denotes the name of the operation that initiated the
lock hold (this will be OTHER for certain uncategorized operations) and
they export the hold time values in nanoseconds.
dfs.namenode.fslock.fair
true
If this is true, the FS Namesystem lock will be used in Fair mode,
which will help to prevent writer threads from being starved, but can provide
lower lock throughput. See java.util.concurrent.locks.ReentrantReadWriteLock
for more information on fair/non-fair locks.
dfs.datanode.lock.fair
true
If this is true, the Datanode FsDataset lock will be used in Fair
mode, which will help to prevent writer threads from being starved, but can
lower lock throughput. See java.util.concurrent.locks.ReentrantReadWriteLock
for more information on fair/non-fair locks.
dfs.namenode.startup.delay.block.deletion.sec
0
The delay in seconds at which we will pause the blocks deletion
after Namenode startup. By default it's disabled.
In the case a directory has large number of directories and files are
deleted, suggested delay is one hour to give the administrator enough time
to notice large number of pending deletion blocks and take corrective
action.
dfs.datanode.block.id.layout.upgrade.threads
6
The number of threads to use when creating hard links from
current to previous blocks during upgrade of a DataNode to block ID-based
block layout (see HDFS-6482 for details on the layout).
dfs.namenode.list.encryption.zones.num.responses
100
When listing encryption zones, the maximum number of zones
that will be returned in a batch. Fetching the list incrementally in
batches improves namenode performance.
dfs.namenode.list.reencryption.status.num.responses
100
When listing re-encryption status, the maximum number of zones
that will be returned in a batch. Fetching the list incrementally in
batches improves namenode performance.
dfs.namenode.list.openfiles.num.responses
1000
When listing open files, the maximum number of open files that will be
returned in a single batch. Fetching the list incrementally in batches
improves namenode performance.
dfs.namenode.edekcacheloader.interval.ms
1000
When KeyProvider is configured, the interval time of warming
up edek cache on NN starts up / becomes active. All edeks will be loaded
from KMS into provider cache. The edek cache loader will try to warm up the
cache until succeed or NN leaves active state.
dfs.namenode.edekcacheloader.initial.delay.ms
3000
When KeyProvider is configured, the time delayed until the first
attempt to warm up edek cache on NN start up / become active.
dfs.namenode.reencrypt.sleep.interval
1m
Interval the re-encrypt EDEK thread sleeps in the main loop. The
interval accepts units. If none given, millisecond is assumed.
dfs.namenode.reencrypt.batch.size
1000
How many EDEKs should the re-encrypt thread process in one batch.
dfs.namenode.reencrypt.throttle.limit.handler.ratio
1.0
Throttling ratio for the re-encryption, indicating what fraction
of time should the re-encrypt handler thread work under NN read lock.
Larger than 1.0 values are interpreted as 1.0. Negative value or 0 are
invalid values and will fail NN startup.
dfs.namenode.reencrypt.throttle.limit.updater.ratio
1.0
Throttling ratio for the re-encryption, indicating what fraction
of time should the re-encrypt updater thread work under NN write lock.
Larger than 1.0 values are interpreted as 1.0. Negative value or 0 are
invalid values and will fail NN startup.
dfs.namenode.reencrypt.edek.threads
10
Maximum number of re-encrypt threads to contact the KMS
and re-encrypt the edeks.
dfs.namenode.inotify.max.events.per.rpc
1000
Maximum number of events that will be sent to an inotify client
in a single RPC response. The default value attempts to amortize away
the overhead for this RPC while avoiding huge memory requirements for the
client and NameNode (1000 events should consume no more than 1 MB.)
dfs.user.home.dir.prefix
/user
The directory to prepend to user name to get the user's
home direcotry.
dfs.datanode.cache.revocation.timeout.ms
900000
When the DFSClient reads from a block file which the DataNode is
caching, the DFSClient can skip verifying checksums. The DataNode will
keep the block file in cache until the client is done. If the client takes
an unusually long time, though, the DataNode may need to evict the block
file from the cache anyway. This value controls how long the DataNode will
wait for the client to release a replica that it is reading without
checksums.
dfs.datanode.cache.revocation.polling.ms
500
How often the DataNode should poll to see if the clients have
stopped using a replica that the DataNode wants to uncache.
dfs.storage.policy.enabled
true
Allow users to change the storage policy on files and directories.
dfs.storage.policy.permissions.superuser-only
false
Allow only superuser role to change the storage policy on files and
directories.
dfs.namenode.legacy-oiv-image.dir
Determines where to save the namespace in the old fsimage format
during checkpointing by standby NameNode or SecondaryNameNode. Users can
dump the contents of the old format fsimage by oiv_legacy command. If
the value is not specified, old format fsimage will not be saved in
checkpoint.
dfs.namenode.top.enabled
true
Enable nntop: reporting top users on namenode
dfs.namenode.top.window.num.buckets
10
Number of buckets in the rolling window implementation of nntop
dfs.namenode.top.num.users
10
Number of top users returned by the top tool
dfs.namenode.top.windows.minutes
1,5,25
comma separated list of nntop reporting periods in minutes
dfs.webhdfs.ugi.expire.after.access
600000
How long in milliseconds after the last access
the cached UGI will expire. With 0, never expire.
dfs.namenode.blocks.per.postponedblocks.rescan
10000
Number of blocks to rescan for each iteration of
postponedMisreplicatedBlocks.
dfs.datanode.block-pinning.enabled
false
Whether pin blocks on favored DataNode.
dfs.client.block.write.locateFollowingBlock.initial.delay.ms
400
The initial delay (unit is ms) for locateFollowingBlock,
the delay time will increase exponentially(double) for each retry
until dfs.client.block.write.locateFollowingBlock.max.delay.ms is reached,
after that the delay for each retry will be
dfs.client.block.write.locateFollowingBlock.max.delay.ms.
dfs.client.block.write.locateFollowingBlock.max.delay.ms
60000
The maximum delay (unit is ms) before retrying locateFollowingBlock.
dfs.ha.zkfc.nn.http.timeout.ms
20000
The HTTP connection and read timeout value (unit is ms ) when DFS ZKFC
tries to get local NN thread dump after local NN becomes
SERVICE_NOT_RESPONDING or SERVICE_UNHEALTHY.
If it is set to zero, DFS ZKFC won't get local NN thread dump.
dfs.ha.nn.not-become-active-in-safemode
false
This will prevent safe mode namenodes to become active or observer while other standby
namenodes might be ready to serve requests when it is set to true.
dfs.ha.tail-edits.in-progress
false
Whether enable standby namenode to tail in-progress edit logs.
Clients might want to turn it on when they want Standby NN to have
more up-to-date data. When using the QuorumJournalManager, this enables
tailing of edit logs via the RPC-based mechanism, rather than streaming,
which allows for much fresher data.
dfs.namenode.state.context.enabled
false
Whether enable namenode sending back its current txnid back to client.
Setting this to true is required by Consistent Read from Standby feature.
But for regular cases, this should be set to false to avoid the overhead
of updating and maintaining this state.
dfs.namenode.ec.system.default.policy
RS-6-3-1024k
The default erasure coding policy name will be used
on the path if no policy name is passed.
dfs.namenode.ec.policies.max.cellsize
4194304
The maximum cell size of erasure coding policy. Default is 4MB.
dfs.namenode.ec.userdefined.policy.allowed
true
If set to false, doesn't allow addition of user defined
erasure coding policies.
dfs.datanode.ec.reconstruction.stripedread.timeout.millis
5000
Datanode striped read timeout in milliseconds.
dfs.datanode.ec.reconstruction.stripedread.buffer.size
65536
Datanode striped read buffer size.
dfs.datanode.ec.reconstruction.threads
8
Number of threads used by the Datanode for background
reconstruction work.
dfs.datanode.ec.reconstruction.xmits.weight
0.5
Datanode uses xmits weight to calculate the relative cost of EC recovery
tasks comparing to replicated block recovery, of which xmits is always 1.
Namenode then uses xmits reported from datanode to throttle recovery tasks
for EC and replicated blocks.
The xmits of an erasure coding recovery task is calculated as the maximum
value between the number of read streams and the number of write streams.
dfs.datanode.ec.reconstruction.validation
false
Decide if datanode validates that EC reconstruction tasks reconstruct
target blocks correctly. When validation fails, reconstruction tasks
will fail and be retried by namenode.
dfs.namenode.quota.init-threads
12
The number of concurrent threads to be used in quota initialization. The
speed of quota initialization also affects the namenode fail-over latency.
If the size of name space is big, try increasing this to 16 or higher.
dfs.datanode.transfer.socket.send.buffer.size
0
Socket send buffer size for DataXceiver (mirroring packets to downstream
in pipeline). This may affect TCP connection throughput.
If it is set to zero or negative value, no buffer size will be set
explicitly, thus enable tcp auto-tuning on some system.
The default value is 0.
dfs.datanode.transfer.socket.recv.buffer.size
0
Socket receive buffer size for DataXceiver (receiving packets from client
during block writing). This may affect TCP connection throughput.
If it is set to zero or negative value, no buffer size will be set
explicitly, thus enable tcp auto-tuning on some system.
The default value is 0.
dfs.namenode.upgrade.domain.factor
${dfs.replication}
This is valid only when block placement policy is set to
BlockPlacementPolicyWithUpgradeDomain. It defines the number of
unique upgrade domains any block's replicas should have.
When the number of replicas is less or equal to this value, the policy
ensures each replica has an unique upgrade domain. When the number of
replicas is greater than this value, the policy ensures the number of
unique domains is at least this value.
dfs.datanode.bp-ready.timeout
20
The maximum wait time for datanode to be ready before failing the
received request. Setting this to 0 fails requests right away if the
datanode is not yet registered with the namenode. This wait time
reduces initial request failures after datanode restart.
Support multiple time unit suffix(case insensitive), as described
in dfs.heartbeat.interval.If no time unit is specified then seconds
is assumed.
dfs.datanode.cached-dfsused.check.interval.ms
600000
The interval check time of loading DU_CACHE_FILE in each volume.
When the cluster doing the rolling upgrade operations, it will
usually lead dfsUsed cache file of each volume expired and redo the
du operations in datanode and that makes datanode start slowly. Adjust
this property can make cache file be available for the time as you want.
dfs.webhdfs.rest-csrf.enabled
false
If true, then enables WebHDFS protection against cross-site request forgery
(CSRF). The WebHDFS client also uses this property to determine whether or
not it needs to send the custom CSRF prevention header in its HTTP requests.
dfs.webhdfs.rest-csrf.custom-header
X-XSRF-HEADER
The name of a custom header that HTTP requests must send when protection
against cross-site request forgery (CSRF) is enabled for WebHDFS by setting
dfs.webhdfs.rest-csrf.enabled to true. The WebHDFS client also uses this
property to determine whether or not it needs to send the custom CSRF
prevention header in its HTTP requests.
dfs.webhdfs.rest-csrf.methods-to-ignore
GET,OPTIONS,HEAD,TRACE
A comma-separated list of HTTP methods that do not require HTTP requests to
include a custom header when protection against cross-site request forgery
(CSRF) is enabled for WebHDFS by setting dfs.webhdfs.rest-csrf.enabled to
true. The WebHDFS client also uses this property to determine whether or
not it needs to send the custom CSRF prevention header in its HTTP requests.
dfs.webhdfs.rest-csrf.browser-useragents-regex
^Mozilla.*,^Opera.*
A comma-separated list of regular expressions used to match against an HTTP
request's User-Agent header when protection against cross-site request
forgery (CSRF) is enabled for WebHDFS by setting
dfs.webhdfs.reset-csrf.enabled to true. If the incoming User-Agent matches
any of these regular expressions, then the request is considered to be sent
by a browser, and therefore CSRF prevention is enforced. If the request's
User-Agent does not match any of these regular expressions, then the request
is considered to be sent by something other than a browser, such as scripted
automation. In this case, CSRF is not a potential attack vector, so
the prevention is not enforced. This helps achieve backwards-compatibility
with existing automation that has not been updated to send the CSRF
prevention header.
dfs.xframe.enabled
true
If true, then enables protection against clickjacking by returning
X_FRAME_OPTIONS header value set to SAMEORIGIN.
Clickjacking protection prevents an attacker from using transparent or
opaque layers to trick a user into clicking on a button
or link on another page.
dfs.xframe.value
SAMEORIGIN
This configration value allows user to specify the value for the
X-FRAME-OPTIONS. The possible values for this field are
DENY, SAMEORIGIN and ALLOW-FROM. Any other value will throw an
exception when namenode and datanodes are starting up.
dfs.balancer.keytab.enabled
false
Set to true to enable login using a keytab for Kerberized Hadoop.
dfs.balancer.address
0.0.0.0:0
The hostname used for a keytab based Kerberos login. Keytab based login
can be enabled with dfs.balancer.keytab.enabled.
dfs.balancer.keytab.file
The keytab file used by the Balancer to login as its
service principal. The principal name is configured with
dfs.balancer.kerberos.principal. Keytab based login can be
enabled with dfs.balancer.keytab.enabled.
dfs.balancer.kerberos.principal
The Balancer principal. This is typically set to
balancer/_HOST@REALM.TLD. The Balancer will substitute _HOST with its
own fully qualified hostname at startup. The _HOST placeholder
allows using the same configuration setting on different servers.
Keytab based login can be enabled with dfs.balancer.keytab.enabled.
dfs.http.client.retry.policy.enabled
false
If "true", enable the retry policy of WebHDFS client.
If "false", retry policy is turned off.
Enabling the retry policy can be quite useful while using WebHDFS to
copy large files between clusters that could timeout, or
copy files between HA clusters that could failover during the copy.
dfs.http.client.retry.policy.spec
10000,6,60000,10
Specify a policy of multiple linear random retry for WebHDFS client,
e.g. given pairs of sleep time and number of retries (t0, n0), (t1, n1),
..., the first n0 retries sleep t0 milliseconds on average,
the following n1 retries sleep t1 milliseconds on average, and so on.
dfs.http.client.failover.max.attempts
15
Specify the max number of failover attempts for WebHDFS client
in case of network exception.
dfs.http.client.retry.max.attempts
10
Specify the max number of retry attempts for WebHDFS client,
if the difference between retried attempts and failovered attempts is
larger than the max number of retry attempts, there will be no more
retries.
dfs.http.client.failover.sleep.base.millis
500
Specify the base amount of time in milliseconds upon which the
exponentially increased sleep time between retries or failovers
is calculated for WebHDFS client.
dfs.http.client.failover.sleep.max.millis
15000
Specify the upper bound of sleep time in milliseconds between
retries or failovers for WebHDFS client.
dfs.namenode.hosts.provider.classname
org.apache.hadoop.hdfs.server.blockmanagement.HostFileManager
The class that provides access for host files.
org.apache.hadoop.hdfs.server.blockmanagement.HostFileManager is used
by default which loads files specified by dfs.hosts and dfs.hosts.exclude.
If org.apache.hadoop.hdfs.server.blockmanagement.CombinedHostFileManager is
used, it will load the JSON file defined in dfs.hosts.
To change class name, nn restart is required. "dfsadmin -refreshNodes" only
refreshes the configuration files used by the class.
datanode.https.port
50475
HTTPS port for DataNode.
dfs.namenode.get-blocks.max-qps
20
The maximum number of getBlocks RPCs data movement utilities can make to
a NameNode per second. Values less than or equal to 0 disable throttling.
This affects anything that uses a NameNodeConnector, i.e., the Balancer,
Mover, and StoragePolicySatisfier.
dfs.namenode.get-blocks.check.operation
true
Set false to disable checkOperation and getBlocks for Balancer
will route to Standby NameNode for HA mode setup.
dfs.balancer.dispatcherThreads
200
Size of the thread pool for the HDFS balancer block mover.
dispatchExecutor
dfs.balancer.movedWinWidth
5400000
Window of time in ms for the HDFS balancer tracking blocks and its
locations.
dfs.balancer.moverThreads
1000
Thread pool size for executing block moves.
moverThreadAllocator
dfs.balancer.max-size-to-move
10737418240
Maximum number of bytes that can be moved by the balancer in a single
thread.
dfs.balancer.getBlocks.min-block-size
10485760
Minimum block threshold size in bytes to ignore when fetching a source's
block list.
dfs.balancer.getBlocks.size
2147483648
Total size in bytes of Datanode blocks to get when fetching a source's
block list.
dfs.balancer.block-move.timeout
0
Maximum amount of time in milliseconds for a block to move. If this is set
greater than 0, Balancer will stop waiting for a block move completion
after this time. In typical clusters, a 3 to 5 minute timeout is reasonable.
If timeout happens to a large proportion of block moves, this needs to be
increased. It could also be that too much work is dispatched and many nodes
are constantly exceeding the bandwidth limit as a result. In that case,
other balancer parameters might need to be adjusted.
It is disabled (0) by default.
dfs.balancer.max-no-move-interval
60000
If this specified amount of time has elapsed and no block has been moved
out of a source DataNode, on more effort will be made to move blocks out of
this DataNode in the current Balancer iteration.
dfs.balancer.max-iteration-time
1200000
Maximum amount of time while an iteration can be run by the Balancer. After
this time the Balancer will stop the iteration, and reevaluate the work
needs to be done to Balance the cluster. The default value is 20 minutes.
dfs.block.invalidate.limit
1000
The maximum number of invalidate blocks sent by namenode to a datanode
per heartbeat deletion command. This property works with
"dfs.namenode.invalidate.work.pct.per.iteration" to throttle block
deletions.
dfs.balancer.service.interval
5m
The schedule interval of balancer when running as a long service.
dfs.balancer.service.retries.on.exception
5
When the balancer is executed as a long-running service, it will retry upon encountering an exception. This
configuration determines how many times it will retry before considering the exception to be fatal and quitting.
dfs.block.misreplication.processing.limit
10000
Maximum number of blocks to process for initializing replication queues.
dfs.block.placement.ec.classname
org.apache.hadoop.hdfs.server.blockmanagement.BlockPlacementPolicyRackFaultTolerant
Placement policy class for striped files.
Defaults to BlockPlacementPolicyRackFaultTolerant.class
dfs.block.replicator.classname
org.apache.hadoop.hdfs.server.blockmanagement.BlockPlacementPolicyDefault
Class representing block placement policy for non-striped files.
There are six block placement policies currently being supported:
BlockPlacementPolicyDefault, BlockPlacementPolicyWithNodeGroup,
BlockPlacementPolicyRackFaultTolerant, BlockPlacementPolicyWithUpgradeDomain,
AvailableSpaceBlockPlacementPolicy and AvailableSpaceRackFaultTolerantBlockPlacementPolicy.
BlockPlacementPolicyDefault chooses the desired number of targets
for placing block replicas in a default way. BlockPlacementPolicyWithNodeGroup
places block replicas on environment with node-group layer. BlockPlacementPolicyRackFaultTolerant
places the replicas to more racks.
BlockPlacementPolicyWithUpgradeDomain places block replicas that honors upgrade domain policy.
AvailableSpaceBlockPlacementPolicy places block replicas based on space balanced policy.
AvailableSpaceRackFaultTolerantBlockPlacementPolicy places block replicas based on
space balanced rack fault tolerant policy.
The details of placing replicas are documented in the javadoc of the corresponding policy classes.
The default policy is BlockPlacementPolicyDefault, and the corresponding class is
org.apache.hadoop.hdfs.server.blockmanagement.BlockPlacementPolicyDefault.
dfs.blockreport.incremental.intervalMsec
0
If set to a positive integer, the value in ms to wait between sending
incremental block reports from the Datanode to the Namenode.
dfs.checksum.type
CRC32C
Checksum type
dfs.checksum.combine.mode
MD5MD5CRC
Defines how lower-level chunk/block checksums are combined into file-level
checksums; the original MD5MD5CRC mode is not comparable between files
with different block layouts, while modes like COMPOSITE_CRC are
comparable independently of block layout.
dfs.checksum.ec.socket-timeout
3000
Default timeout value in milliseconds for computing the checksum of striped blocks.
Recommended to set the same value between client and DNs in a cluster because mismatching
may cause exhausting handler threads.
dfs.client.block.write.locateFollowingBlock.retries
5
Number of retries to use when finding the next block during HDFS writes.
dfs.client.failover.proxy.provider
The prefix (plus a required nameservice ID) for the class name of the configured
Failover proxy provider for the host. For normal HA mode, please consult
the "Configuration Details" section of the HDFS High Availability documentation.
For observer reading mode, please choose a custom class--ObserverReadProxyProvider.
dfs.client.failover.random.order
true
Determines if the failover proxies are picked in random order instead of the
configured order. Random order may be enabled for better load balancing
or to avoid always hitting failed ones first if the failed ones appear in the
beginning of the configured or resolved list.
For example, In the case of multiple RBF routers or ObserverNameNodes,
it is recommended to be turned on for load balancing.
The config name can be extended with an optional nameservice ID
(of form dfs.client.failover.random.order[.nameservice]) in case multiple
nameservices exist and random order should be enabled for specific
nameservices.
dfs.client.failover.resolve-needed
false
Determines if the given nameservice address is a domain name which needs to
be resolved (using the resolver configured by dfs.client.failover.resolver-impl).
This adds a transparency layer in the client so physical server address
can change without changing the client. The config name can be extended with
an optional nameservice ID (of form dfs.client.failover.resolve-needed[.nameservice])
to configure specific nameservices when multiple nameservices exist.
dfs.client.failover.resolver.impl
org.apache.hadoop.net.DNSDomainNameResolver
Determines what class to use to resolve nameservice name to specific machine
address(es). The config name can be extended with an optional nameservice ID
(of form dfs.client.failover.resolver.impl[.nameservice]) to configure
specific nameservices when multiple nameservices exist.
dfs.client.failover.resolver.useFQDN
true
Determines whether the resolved result is fully qualified domain name instead
of pure IP address(es). The config name can be extended with an optional
nameservice ID (of form dfs.client.failover.resolver.impl[.nameservice]) to
configure specific nameservices when multiple nameservices exist.
In secure environment, this has to be enabled since Kerberos is using fqdn
in machine's principal therefore accessing servers by IP won't be recognized
by the KDC.
dfs.client.key.provider.cache.expiry
864000000
DFS client security key cache expiration in milliseconds.
dfs.client.max.block.acquire.failures
3
Maximum failures allowed when trying to get block information from a specific datanode.
dfs.client.read.prefetch.size
The number of bytes for the DFSClient will fetch from the Namenode
during a read operation. Defaults to 10 * ${dfs.blocksize}.
dfs.client.read.uri.cache.enabled
false
If true, dfs client will use cache when creating URI based on host:port
to reduce the frequency of URI object creation.
dfs.client.read.short.circuit.replica.stale.threshold.ms
1800000
Threshold in milliseconds for read entries during short-circuit local reads.
dfs.client.read.shortcircuit.buffer.size
1048576
Buffer size in bytes for short-circuit local reads.
dfs.client.short.circuit.num
1
Number of short-circuit caches. This setting should
be in the range 1 - 5. Lower values will result in lower CPU consumption; higher
values may speed up massive parallel reading files.
dfs.client.read.striped.threadpool.size
18
The maximum number of threads used for parallel reading
in striped layout.
dfs.client.replica.accessor.builder.classes
Comma-separated classes for building ReplicaAccessor. If the classes
are specified, client will use external BlockReader that uses the
ReplicaAccessor built by the builder.
dfs.client.retry.interval-ms.get-last-block-length
4000
Retry interval in milliseconds to wait between retries in getting
block lengths from the datanodes.
dfs.client.retry.max.attempts
10
Max retry attempts for DFSClient talking to namenodes.
dfs.client.retry.policy.enabled
false
If true, turns on DFSClient retry policy.
dfs.client.retry.policy.spec
10000,6,60000,10
Specify a policy of multiple linear random retry for the DFS client,
e.g. given pairs of sleep time and number of retries (t0, n0), (t1, n1),
..., the first n0 retries sleep t0 milliseconds on average,
the following n1 retries sleep t1 milliseconds on average, and so on.
dfs.client.retry.times.get-last-block-length
3
Number of retries for calls to fetchLocatedBlocksAndGetLastBlockLength().
dfs.client.retry.window.base
3000
Base time window in ms for DFSClient retries. For each retry attempt,
this value is extended linearly (e.g. 3000 ms for first attempt and
first retry, 6000 ms for second retry, 9000 ms for third retry, etc.).
dfs.client.pipeline.recovery.max-retries
5
if the DFS client encounters errors in write pipeline,
retry up to the number defined by this property before giving up.
dfs.client.socket-timeout
60000
Default timeout value in milliseconds for all sockets.
dfs.client.socketcache.capacity
16
Socket cache capacity (in entries) for short-circuit reads.
If this value is set to 0, the client socket cache is disabled.
dfs.client.socketcache.expiryMsec
3000
Socket cache expiration for short-circuit reads in msec.
dfs.client.test.drop.namenode.response.number
0
The number of Namenode responses dropped by DFSClient for each RPC call. Used
for testing the NN retry cache.
dfs.client.hedged.read.threadpool.size
0
Support 'hedged' reads in DFSClient. To enable this feature, set the parameter
to a positive number. The threadpool size is how many threads to dedicate
to the running of these 'hedged', concurrent reads in your client.
dfs.client.hedged.read.threshold.millis
500
Configure 'hedged' reads in DFSClient. This is the number of milliseconds
to wait before starting up a 'hedged' read.
dfs.client.write.byte-array-manager.count-limit
2048
The maximum number of arrays allowed for each array length.
dfs.client.write.byte-array-manager.count-reset-time-period-ms
10000
The time period in milliseconds that the allocation count for each array length is
reset to zero if there is no increment.
dfs.client.write.byte-array-manager.count-threshold
128
The count threshold for each array length so that a manager is created only after the
allocation count exceeds the threshold. In other words, the particular array length
is not managed until the allocation count exceeds the threshold.
dfs.client.write.byte-array-manager.enabled
false
If true, enables byte array manager used by DFSOutputStream.
dfs.client.write.max-packets-in-flight
80
The maximum number of DFSPackets allowed in flight.
dfs.client.block.reader.remote.buffer.size
512
The output stream buffer size of a DFSClient remote read. The buffer default value is 512B. The buffer includes
only some request parameters that are: block, blockToken, clientName, startOffset, len, verifyChecksum,
cachingStrategy.
dfs.content-summary.limit
5000
The maximum content summary counts allowed in one locking period. 0 or a negative number
means no limit (i.e. no yielding).
dfs.content-summary.sleep-microsec
500
The length of time in microseconds to put the thread to sleep, between reaquiring the locks
in content summary computation.
dfs.data.transfer.client.tcpnodelay
true
If true, set TCP_NODELAY to sockets for transferring data from DFS client.
dfs.data.transfer.server.tcpnodelay
true
If true, set TCP_NODELAY to sockets for transferring data between Datanodes.
dfs.data.transfer.max.packet.size
16777216
The max size of any single packet.
dfs.datanode.balance.max.concurrent.moves
100
Maximum number of threads for Datanode balancer pending moves. This
value is reconfigurable via the "dfsadmin -reconfig" command.
dfs.datanode.data.transfer.bandwidthPerSec
0
Specifies the maximum amount of bandwidth that the data transfering can utilize for transfering block when
BlockConstructionStage is
PIPELINE_SETUP_CREATE and clientName is empty.
When the bandwidth value is zero, there is no limit.
dfs.datanode.data.write.bandwidthPerSec
0
Specifies the maximum amount of bandwidth that the data transfering can utilize for writing block or pipeline
recovery when
BlockConstructionStage is PIPELINE_SETUP_APPEND_RECOVERY or PIPELINE_SETUP_STREAMING_RECOVERY.
When the bandwidth value is zero, there is no limit.
dfs.datanode.ec.reconstruct.read.bandwidthPerSec
0
Specifies the maximum amount of bandwidth that the EC reconstruction can utilize for reading.
When the bandwidth value is zero, there is no limit.
dfs.datanode.ec.reconstruct.write.bandwidthPerSec
0
Specifies the maximum amount of bandwidth that the EC reconstruction can utilize for writing.
When the bandwidth value is zero, there is no limit.
dfs.datanode.fsdataset.factory
The class name for the underlying storage that stores replicas for a
Datanode. Defaults to
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.FsDatasetFactory.
dfs.datanode.fsdataset.volume.choosing.policy
The class name of the policy for choosing volumes in the list of
directories. Defaults to
org.apache.hadoop.hdfs.server.datanode.fsdataset.RoundRobinVolumeChoosingPolicy.
If you would like to take into account available disk space, set the
value to
"org.apache.hadoop.hdfs.server.datanode.fsdataset.AvailableSpaceVolumeChoosingPolicy".
dfs.datanode.hostname
Optional. The hostname for the Datanode containing this
configuration file. Will be different for each machine.
Defaults to current hostname.
dfs.datanode.lazywriter.interval.sec
60
Interval in seconds for Datanodes for lazy persist writes.
dfs.datanode.network.counts.cache.max.size
2147483647
The maximum number of entries the datanode per-host network error
count cache may contain.
dfs.datanode.oob.timeout-ms
1500,0,0,0
Timeout value when sending OOB response for each OOB type, which are
OOB_RESTART, OOB_RESERVED1, OOB_RESERVED2, and OOB_RESERVED3,
respectively. Currently, only OOB_RESTART is used.
dfs.datanode.parallel.volumes.load.threads.num
Maximum number of threads to use for upgrading data directories.
The default value is the number of storage directories in the
DataNode.
dfs.datanode.ram.disk.replica.tracker
Name of the class implementing the RamDiskReplicaTracker interface.
Defaults to
org.apache.hadoop.hdfs.server.datanode.fsdataset.impl.RamDiskReplicaLruTracker.
dfs.datanode.restart.replica.expiration
50
During shutdown for restart, the amount of time in seconds budgeted for
datanode restart.
dfs.datanode.socket.reuse.keepalive
4000
The window of time in ms before the DataXceiver closes a socket for a
single request. If a second request occurs within that window, the
socket can be reused.
dfs.datanode.socket.write.timeout
480000
Timeout in ms for clients socket writes to DataNodes.
dfs.datanode.sync.behind.writes.in.background
false
If set to true, then sync_file_range() system call will occur
asynchronously. This property is only valid when the property
dfs.datanode.sync.behind.writes is true.
dfs.datanode.transferTo.allowed
true
If false, break block transfers on 32-bit machines greater than
or equal to 2GB into smaller chunks.
dfs.datanode.fixed.volume.size
false
If false, call function getTotalSpace of File to get capacity of volume
during every heartbeat.
If true, cache the capacity when when the first call, and reuse it later.
dfs.datanode.replica.cache.root.dir
Use this key to change root dir of replica cache.
The default root dir is currentDir.
dfs.datanode.replica.cache.expiry.time
5m
Living time of replica cached files in milliseconds.
dfs.ha.fencing.methods
A list of scripts or Java classes which will be used to fence
the Active NameNode during a failover. See the HDFS High
Availability documentation for details on automatic HA
configuration.
dfs.ha.standby.checkpoints
true
If true, a NameNode in Standby state periodically takes a checkpoint
of the namespace, saves it to its local storage and then upload to
the remote NameNode.
dfs.ha.zkfc.port
8019
The port number that the zookeeper failover controller RPC
server binds to.
dfs.ha.allow.stale.reads
false
If true, a NameNode in Standby state can process read request and the result
could be stale.
dfs.journalnode.edits.dir
/tmp/hadoop/dfs/journalnode/
The directory where the journal edit files are stored.
dfs.journalnode.enable.sync
true
If true, the journal nodes wil sync with each other. The journal nodes
will periodically gossip with other journal nodes to compare edit log
manifests and if they detect any missing log segment, they will download
it from the other journal nodes.
dfs.journalnode.sync.interval
120000
Time interval, in milliseconds, between two Journal Node syncs.
This configuration takes effect only if the journalnode sync is enabled
by setting the configuration parameter dfs.journalnode.enable.sync to true.
dfs.journalnode.edit-cache-size.bytes
The size, in bytes, of the in-memory cache of edits to keep on the
JournalNode. This cache is used to serve edits for tailing via the RPC-based
mechanism, and is only enabled when dfs.ha.tail-edits.in-progress is true.
Transactions range in size but are around 200 bytes on average, so the
default of 1MB can store around 5000 transactions.
dfs.journalnode.edit-cache-size.fraction
0.5f
This ratio refers to the proportion of the maximum memory of the JVM.
Used to calculate the size of the edits cache that is kept in the JournalNode's memory.
This config is an alternative to the dfs.journalnode.edit-cache-size.bytes.
And it is used to serve edits for tailing via the RPC-based mechanism, and is only
enabled when dfs.ha.tail-edits.in-progress is true. Transactions range in size but
are around 200 bytes on average, so the default of 1MB can store around 5000 transactions.
So we can configure a reasonable value based on the maximum memory. The recommended value
is less than 0.9. If we set dfs.journalnode.edit-cache-size.bytes, this parameter will
not take effect.
dfs.journalnode.kerberos.internal.spnego.principal
Kerberos SPNEGO principal name used by the journal node.
dfs.journalnode.kerberos.principal
Kerberos principal name for the journal node.
dfs.journalnode.keytab.file
Kerberos keytab file for the journal node.
dfs.batched.ls.limit
100
Limit the number of paths that can be listed in a single batched
listing call. printed by ls. If less or equal to
zero, at most DFS_LIST_LIMIT_DEFAULT (= 1000) will be printed.
dfs.ls.limit
1000
Limit the number of files printed by ls. If less or equal to
zero, at most DFS_LIST_LIMIT_DEFAULT (= 1000) will be printed.
dfs.mover.movedWinWidth
5400000
The minimum time interval, in milliseconds, that a block can be
moved to another location again.
dfs.mover.moverThreads
1000
Configure the balancer's mover thread pool size.
dfs.mover.retry.max.attempts
10
The maximum number of retries before the mover consider the
move failed.
dfs.mover.keytab.enabled
false
Set to true to enable login using a keytab for Kerberized Hadoop.
dfs.mover.address
0.0.0.0:0
The hostname used for a keytab based Kerberos login. Keytab based login
can be enabled with dfs.mover.keytab.enabled.
dfs.mover.keytab.file
The keytab file used by the Mover to login as its
service principal. The principal name is configured with
dfs.mover.kerberos.principal. Keytab based login can be
enabled with dfs.mover.keytab.enabled.
dfs.mover.kerberos.principal
The Mover principal. This is typically set to
mover/_HOST@REALM.TLD. The Mover will substitute _HOST with its
own fully qualified hostname at startup. The _HOST placeholder
allows using the same configuration setting on different servers.
Keytab based login can be enabled with dfs.mover.keytab.enabled.
dfs.mover.max-no-move-interval
60000
If this specified amount of time has elapsed and no block has been moved
out of a source DataNode, on more effort will be made to move blocks out of
this DataNode in the current Mover iteration.
dfs.namenode.audit.log.token.tracking.id
false
If true, adds a tracking ID for all audit log events.
dfs.namenode.audit.log.with.remote.port
false
If true, adds a port of RPC call to callerContext for all audit log events.
dfs.namenode.available-space-block-placement-policy.balanced-space-preference-fraction
0.6
Only used when the dfs.block.replicator.classname is set to
org.apache.hadoop.hdfs.server.blockmanagement.AvailableSpaceBlockPlacementPolicy.
Special value between 0 and 1, noninclusive. Increases chance of
placing blocks on Datanodes with less disk space used.
dfs.namenode.available-space-block-placement-policy.balanced-space-tolerance
5
Only used when the dfs.block.replicator.classname is set to
org.apache.hadoop.hdfs.server.blockmanagement.AvailableSpaceBlockPlacementPolicy.
Special value between 0 and 20, inclusive. if the value is set beyond the scope,
this value will be set as 5 by default, Increases tolerance of
placing blocks on Datanodes with similar disk space used.
dfs.namenode.available-space-block-placement-policy.balance-local-node
false
Only used when the dfs.block.replicator.classname is set to
org.apache.hadoop.hdfs.server.blockmanagement.AvailableSpaceBlockPlacementPolicy.
If true, balances the local node too.
dfs.namenode.available-space-rack-fault-tolerant-block-placement-policy.balanced-space-preference-fraction
0.6
Only used when the dfs.block.replicator.classname is set to
org.apache.hadoop.hdfs.server.blockmanagement.AvailableSpaceRackFaultTolerantBlockPlacementPolicy.
Special value between 0 and 1, noninclusive. Increases chance of
placing blocks on Datanodes with less disk space used. More the value near 1
more are the chances of choosing the datanode with less percentage of data.
Similarly as the value moves near 0, the chances of choosing datanode with
high load increases as the value reaches near 0.
dfs.namenode.available-space-rack-fault-tolerant-block-placement-policy.balanced-space-tolerance
5
Only used when the dfs.block.replicator.classname is set to
org.apache.hadoop.hdfs.server.blockmanagement.AvailableSpaceRackFaultTolerantBlockPlacementPolicy.
Special value between 0 and 20, inclusive. if the value is set beyond the scope,
this value will be set as 5 by default, Increases tolerance of
placing blocks on Datanodes with similar disk space used.
dfs.namenode.backup.dnrpc-address
Service RPC address for the backup Namenode.
dfs.namenode.delegation.token.always-use
false
For testing. Setting to true always allows the DT secret manager
to be used, even if security is disabled.
dfs.namenode.edits.asynclogging
true
If set to true, enables asynchronous edit logs in the Namenode. If set
to false, the Namenode uses the traditional synchronous edit logs.
dfs.namenode.edits.asynclogging.pending.queue.size
4096
The queue size of edit pending queue for FSEditLogAsync.
dfs.namenode.edits.dir.minimum
1
dfs.namenode.edits.dir includes both required directories
(specified by dfs.namenode.edits.dir.required) and optional directories.
The number of usable optional directories must be greater than or equal
to this property. If the number of usable optional directories falls
below dfs.namenode.edits.dir.minimum, HDFS will issue an error.
This property defaults to 1.
dfs.namenode.edits.journal-plugin
When FSEditLog is creating JournalManagers from dfs.namenode.edits.dir,
and it encounters a URI with a schema different to "file" it loads the
name of the implementing class from
"dfs.namenode.edits.journal-plugin.[schema]". This class must implement
JournalManager and have a constructor which takes (Configuration, URI).
dfs.namenode.file.close.num-committed-allowed
0
Normally a file can only be closed with all its blocks are committed.
When this value is set to a positive integer N, a file can be closed
when N blocks are committed and the rest complete. In case of Erasure Coded
blocks, the committed block shall be allowed only when the block group is
complete. i.e no missing/lost block in the blockgroup.
dfs.namenode.inode.attributes.provider.class
Name of class to use for delegating HDFS authorization.
dfs.namenode.inode.attributes.provider.bypass.users
A list of user principals (in secure cluster) or user names (in insecure
cluster) for whom the external attributes provider will be bypassed for all
operations. This means file attributes stored in HDFS instead of the
external provider will be used for permission checking and be returned when
requested.
dfs.namenode.max-num-blocks-to-log
1000
Puts a limit on the number of blocks printed to the log by the Namenode
after a block report.
dfs.namenode.max.op.size
52428800
Maximum opcode size in bytes.
dfs.namenode.missing.checkpoint.periods.before.shutdown
3
The number of checkpoint period windows (as defined by the property
dfs.namenode.checkpoint.period) allowed by the Namenode to perform
saving the namespace before shutdown.
dfs.namenode.name.cache.threshold
10
Frequently accessed files that are accessed more times than this
threshold are cached in the FSDirectory nameCache.
dfs.namenode.replication.max-streams
2
Hard limit for the number of replication streams other than those with highest-priority.
dfs.namenode.replication.max-streams-hard-limit
4
Hard limit for all replication streams.
dfs.namenode.reconstruction.pending.timeout-sec
300
Timeout in seconds for block reconstruction. If this value is 0 or less,
then it will default to 5 minutes.
dfs.namenode.stale.datanode.minimum.interval
3
Minimum number of missed heartbeats intervals for a datanode to
be marked stale by the Namenode. The actual interval is calculated as
(dfs.namenode.stale.datanode.minimum.interval * dfs.heartbeat.interval)
in seconds. If this value is greater than the property
dfs.namenode.stale.datanode.interval, then the calculated value above
is used.
dfs.namenode.remove.dead.datanode.batchnum
10
Maximum number of datanodes removed by HeartbeatManager per scan.
dfs.namenode.snapshot.capture.openfiles
false
If true, snapshots taken will have an immutable shared copy of
the open files that have valid leases. Even after the open files
grow or shrink in size, snapshot will always have the previous
point-in-time version of the open files, just like all other
closed files. Default is false.
Note: The file length captured for open files in snapshot is
whats recorded in NameNode at the time of snapshot and it may
be shorter than what the client has written till then. In order
to capture the latest length, the client can call hflush/hsync
with the flag SyncFlag.UPDATE_LENGTH on the open files handles.
dfs.namenode.snapshot.skip.capture.accesstime-only-change
false
If accessTime of a file/directory changed but there is no other
modification made to the file/directory, the changed accesstime will
not be captured in next snapshot. However, if there is other modification
made to the file/directory, the latest access time will be captured
together with the modification in next snapshot.
dfs.namenode.snapshotdiff.allow.snap-root-descendant
true
If enabled, snapshotDiff command can be run for any descendant directory
under a snapshot root directory and the diff calculation will be scoped
to the given descendant directory. Otherwise, snapshot diff command can
only be run for a snapshot root directory.
dfs.namenode.snapshotdiff.listing.limit
1000
Limit the number of entries generated by getSnapshotDiffReportListing within
one rpc call to the namenode.If less or equal to zero, at most
DFS_NAMENODE_SNAPSHOT_DIFF_LISTING_LIMIT_DEFAULT (= 1000) will be sent
across to the client within one rpc call.
dfs.namenode.snapshot.max.limit
65536
Limits the maximum number of snapshots allowed per snapshottable
directory.If the configuration is not set, the default limit
for maximum no of snapshots allowed is 65536.
dfs.namenode.snapshot.filesystem.limit
65536
Limits the maximum number of snapshots allowed on the entire filesystem.
If the configuration is not set, the default limit
for maximum no of snapshots allowed is 65536.
dfs.namenode.snapshot.skiplist.max.levels
0
Maximum no of the skip levels to be maintained in the skip list for
storing directory snapshot diffs. By default, it is set to 0 and a linear
list will be used to store the directory snapshot diffs.
dfs.namenode.snapshot.skiplist.interval
10
The interval after which the skip levels will be formed in the skip list
for storing directory snapshot diffs. By default, value is set to 10.
dfs.storage.policy.satisfier.mode
none
Following values are supported - external, none.
If external, StoragePolicySatisfier will be enabled and started as an independent service outside namenode.
If none, StoragePolicySatisfier is disabled.
By default, StoragePolicySatisfier is disabled.
Administrator can dynamically change StoragePolicySatisfier mode by using reconfiguration option.
Dynamic mode change can be achieved in the following way.
1. Edit/update this configuration property values in hdfs-site.xml
2. Execute the reconfig command on hadoop command line prompt.
For example:$hdfs -reconfig namenode nn_host:port start
dfs.storage.policy.satisfier.queue.limit
1000
Storage policy satisfier queue size. This queue contains the currently
scheduled file's inode ID for statisfy the policy.
Default value is 1000.
dfs.storage.policy.satisfier.work.multiplier.per.iteration
1
*Note*: Advanced property. Change with caution.
This determines the total amount of block transfers to begin in
one iteration, for satisfy the policy. The actual number is obtained by
multiplying this multiplier with the total number of live nodes in the
cluster. The result number is the number of blocks to begin transfers
immediately. This number can be any positive, non-zero integer.
dfs.storage.policy.satisfier.recheck.timeout.millis
60000
Blocks storage movements monitor re-check interval in milliseconds.
This check will verify whether any blocks storage movement results arrived from DN
and also verify if any of file blocks movements not at all reported to DN
since dfs.storage.policy.satisfier.self.retry.timeout.
The default value is 1 * 60 * 1000 (1 mins)
dfs.storage.policy.satisfier.self.retry.timeout.millis
300000
If any of file related block movements not at all reported by datanode,
then after this timeout(in milliseconds), the item will be added back to movement needed list
at namenode which will be retried for block movements.
The default value is 5 * 60 * 1000 (5 mins)
dfs.storage.policy.satisfier.retry.max.attempts
3
Max retry to satisfy the block storage policy. After this retry block will be removed
from the movement needed queue.
dfs.storage.policy.satisfier.move.task.retry.max.attempts
3
Max retries for moving task to satisfy the block storage policy.
dfs.storage.policy.satisfier.datanode.cache.refresh.interval.ms
300000
How often to refresh the datanode storages cache in milliseconds. This cache
keeps live datanode storage reports fetched from namenode. After elapsed time,
it will again fetch latest datanodes from namenode.
By default, this parameter is set to 5 minutes.
dfs.storage.policy.satisfier.max.outstanding.paths
10000
Defines the maximum number of paths to satisfy that can be queued up in the
Satisfier call queue in a period of time. Default value is 10000.
dfs.storage.policy.satisfier.address
0.0.0.0:0
The hostname used for a keytab based Kerberos login. Keytab based login
is required when dfs.storage.policy.satisfier.mode is external.
dfs.storage.policy.satisfier.keytab.file
The keytab file used by external StoragePolicySatisfier to login as its
service principal. The principal name is configured with
dfs.storage.policy.satisfier.kerberos.principal. Keytab based login
is required when dfs.storage.policy.satisfier.mode is external.
dfs.storage.policy.satisfier.kerberos.principal
The StoragePolicySatisfier principal. This is typically set to
satisfier/_HOST@REALM.TLD. The StoragePolicySatisfier will substitute
_HOST with its own fully qualified hostname at startup. The _HOST placeholder
allows using the same configuration setting on different servers. Keytab
based login is required when dfs.storage.policy.satisfier.mode is external.
dfs.pipeline.ecn
false
If true, allows ECN (explicit congestion notification) from the
Datanode.
dfs.pipeline.slownode
false
If true, allows slownode information to be replied to Client via PipelineAck.
dfs.qjournal.accept-recovery.timeout.ms
120000
Quorum timeout in milliseconds during accept phase of
recovery/synchronization for a specific segment.
dfs.qjournal.finalize-segment.timeout.ms
120000
Quorum timeout in milliseconds during finalizing for a specific
segment.
dfs.qjournal.get-journal-state.timeout.ms
120000
Timeout in milliseconds when calling getJournalState().
JournalNodes.
dfs.qjournal.new-epoch.timeout.ms
120000
Timeout in milliseconds when getting an epoch number for write
access to JournalNodes.
dfs.qjournal.prepare-recovery.timeout.ms
120000
Quorum timeout in milliseconds during preparation phase of
recovery/synchronization for a specific segment.
dfs.qjournal.queued-edits.limit.mb
10
Queue size in MB for quorum journal edits.
dfs.qjournal.select-input-streams.timeout.ms
20000
Timeout in milliseconds for accepting streams from JournalManagers.
dfs.qjournal.start-segment.timeout.ms
20000
Quorum timeout in milliseconds for starting a log segment.
dfs.qjournal.write-txns.timeout.ms
20000
Write timeout in milliseconds when writing to a quorum of remote
journals.
dfs.qjournal.http.open.timeout.ms
60000
Timeout in milliseconds when open a new HTTP connection to remote
journals.
dfs.qjournal.http.read.timeout.ms
60000
Timeout in milliseconds when reading from a HTTP connection from remote
journals.
dfs.qjournal.parallel-read.num-threads
5
Number of threads per JN to be used for tailing edits.
dfs.quota.by.storage.type.enabled
true
If true, enables quotas based on storage type.
dfs.secondary.namenode.kerberos.principal
Kerberos principal name for the Secondary NameNode.
dfs.secondary.namenode.keytab.file
Kerberos keytab file for the Secondary NameNode.
dfs.web.authentication.simple.anonymous.allowed
If true, allow anonymous user to access WebHDFS. Set to
false to disable anonymous authentication.
dfs.web.ugi
dfs.web.ugi is deprecated. Use hadoop.http.staticuser.user instead.
dfs.webhdfs.netty.high.watermark
65535
High watermark configuration to Netty for Datanode WebHdfs.
dfs.webhdfs.netty.low.watermark
32768
Low watermark configuration to Netty for Datanode WebHdfs.
dfs.webhdfs.oauth2.access.token.provider
Access token provider class for WebHDFS using OAuth2.
Defaults to org.apache.hadoop.hdfs.web.oauth2.ConfCredentialBasedAccessTokenProvider.
dfs.webhdfs.oauth2.client.id
Client id used to obtain access token with either credential or
refresh token.
dfs.webhdfs.oauth2.enabled
false
If true, enables OAuth2 in WebHDFS
dfs.webhdfs.oauth2.refresh.url
URL against which to post for obtaining bearer token with
either credential or refresh token.
ssl.server.keystore.keypassword
Keystore key password for HTTPS SSL configuration
ssl.server.keystore.location
Keystore location for HTTPS SSL configuration
ssl.server.keystore.password
Keystore password for HTTPS SSL configuration
ssl.server.truststore.location
Truststore location for HTTPS SSL configuration
ssl.server.truststore.password
Truststore password for HTTPS SSL configuration
dfs.disk.balancer.max.disk.throughputInMBperSec
10
Maximum disk bandwidth used by diskbalancer
during read from a source disk. The unit is MB/sec.
dfs.disk.balancer.block.tolerance.percent
10
When a disk balancer copy operation is proceeding, the datanode is still
active. So it might not be possible to move the exactly specified
amount of data. So tolerance allows us to define a percentage which
defines a good enough move.
dfs.disk.balancer.max.disk.errors
5
During a block move from a source to destination disk, we might
encounter various errors. This defines how many errors we can tolerate
before we declare a move between 2 disks (or a step) has failed.
dfs.disk.balancer.plan.valid.interval
1d
Maximum amount of time disk balancer plan is valid. This setting
supports multiple time unit suffixes as described in
dfs.heartbeat.interval. If no suffix is specified then milliseconds
is assumed.
dfs.disk.balancer.enabled
true
This enables the diskbalancer feature on a cluster. By default, disk
balancer is enabled.
dfs.disk.balancer.plan.threshold.percent
10
The percentage threshold value for volume Data Density in a plan.
If the absolute value of volume Data Density which is out of
threshold value in a node, it means that the volumes corresponding to
the disks should do the balancing in the plan. The default value is 10.
dfs.namenode.provided.enabled
false
Enables the Namenode to handle provided storages.
dfs.provided.storage.id
DS-PROVIDED
The storage ID used for provided stores.
dfs.provided.aliasmap.class
org.apache.hadoop.hdfs.server.common.blockaliasmap.impl.TextFileRegionAliasMap
The class that is used to specify the input format of the blocks on
provided storages. The default is
org.apache.hadoop.hdfs.server.common.blockaliasmap.impl.TextFileRegionAliasMap which uses
file regions to describe blocks. The file regions are specified as a
delimited text file. Each file region is a 6-tuple containing the
block id, remote file path, offset into file, length of block, the
block pool id containing the block, and the generation stamp of the
block.
dfs.provided.aliasmap.inmemory.batch-size
500
The batch size when iterating over the database backing the aliasmap
dfs.provided.aliasmap.inmemory.dnrpc-address
The address where the aliasmap server will be running. In the case of
HA/Federation where multiple namenodes exist, and if the Namenode is
configured to run the aliasmap server
(dfs.provided.aliasmap.inmemory.enabled is set to true),
the name service id is added to the name, e.g.,
dfs.provided.aliasmap.inmemory.rpc.address.EXAMPLENAMESERVICE.
The value of this property will take the form of host:rpc-port.
dfs.provided.aliasmap.inmemory.rpc.bind-host
The actual address the in-memory aliasmap server will bind to.
If this optional address is set, it overrides the hostname portion of
dfs.provided.aliasmap.inmemory.rpc.address.
This is useful for making the name node listen on all interfaces by
setting it to 0.0.0.0.
dfs.provided.aliasmap.inmemory.leveldb.dir
/tmp
The directory where the leveldb files will be kept
dfs.provided.aliasmap.inmemory.enabled
false
Don't use the aliasmap by default. Some tests will fail
because they try to start the namenode twice with the
same parameters if you turn it on.
dfs.provided.aliasmap.inmemory.server.log
false
Ensures that InMemoryAliasMap server logs every call to it.
Set to false by default.
dfs.provided.aliasmap.text.delimiter
,
The delimiter used when the provided block map is specified as
a text file.
dfs.provided.aliasmap.text.read.file
The path specifying the provided block map as a text file, specified as
a URI.
dfs.provided.aliasmap.text.codec
The codec used to de-compress the provided block map.
dfs.provided.aliasmap.text.write.dir
The path to which the provided block map should be written as a text
file, specified as a URI.
dfs.provided.aliasmap.leveldb.path
The read/write path for the leveldb-based alias map
(org.apache.hadoop.hdfs.server.common.blockaliasmap.impl.LevelDBFileRegionAliasMap).
The path has to be explicitly configured when this alias map is used.
dfs.provided.acls.import.enabled
false
Set to true to inherit ACLs (Access Control Lists) from remote stores
during mount. Disabled by default, i.e., ACLs are not inherited from
remote stores. Note had HDFS ACLs have to be enabled
(dfs.namenode.acls.enabled must be set to true) for this to take effect.
dfs.provided.aliasmap.load.retries
0
The number of retries on the Datanode to load the provided aliasmap;
defaults to 0.
dfs.lock.suppress.warning.interval
10s
Instrumentation reporting long critical sections will suppress
consecutive warnings within this interval.
httpfs.buffer.size
4096
The size buffer to be used when creating or opening httpfs filesystem IO stream.
dfs.webhdfs.use.ipc.callq
true
Enables routing of webhdfs calls through rpc
call queue
dfs.datanode.disk.check.min.gap
15m
The minimum gap between two successive checks of the same DataNode
volume. This setting supports multiple time unit suffixes as described
in dfs.heartbeat.interval. If no suffix is specified then milliseconds
is assumed.
dfs.datanode.disk.check.timeout
10m
Maximum allowed time for a disk check to complete. If the check does not
complete within this time interval then the disk is declared as failed.
This setting supports multiple time unit suffixes as described in
dfs.heartbeat.interval. If no suffix is specified then milliseconds is assumed.
dfs.use.dfs.network.topology
true
Enables DFSNetworkTopology to choose nodes for placing replicas.
When enabled, NetworkTopology will be instantiated as class defined in
property dfs.net.topology.impl, otherwise NetworkTopology will be
instantiated as class defined in property net.topology.impl.
dfs.net.topology.impl
org.apache.hadoop.hdfs.net.DFSNetworkTopology
The implementation class of NetworkTopology used in HDFS. By default,
the class org.apache.hadoop.hdfs.net.DFSNetworkTopology is specified and
used in block placement.
This property only works when dfs.use.dfs.network.topology is true.
dfs.qjm.operations.timeout
60s
Common key to set timeout for related operations in
QuorumJournalManager. This setting supports multiple time unit suffixes
as described in dfs.heartbeat.interval.
If no suffix is specified then milliseconds is assumed.
dfs.reformat.disabled
false
Disable reformat of NameNode. If it's value is set to "true"
and metadata directories already exist then attempt to format NameNode
will throw NameNodeFormatException.
dfs.namenode.block.deletion.lock.threshold.ms
50
The limit of single time lock holding duration for the block asynchronous
deletion thread.
dfs.namenode.block.deletion.unlock.interval.ms
10
The sleep interval for yield lock.
When the single time lock holding duration of the block asynchronous deletion
thread exceeds limit, sleeping to yield lock.
dfs.namenode.rpc-address.auxiliary-ports
A comma separated list of auxiliary ports for the NameNode to listen on.
This allows exposing multiple NN addresses to clients.
Particularly, it is used to enforce different SASL levels on different ports.
Empty list indicates that auxiliary ports are disabled.
dfs.namenode.send.qop.enabled
false
A boolean specifies whether NameNode should encrypt the established QOP
and include it in block token. The encrypted QOP will be used by DataNode
as target QOP, overwriting DataNode configuration. This ensures DataNode
will use exactly the same QOP NameNode and client has already agreed on.
dfs.encrypt.data.overwrite.downstream.derived.qop
false
A boolean specifies whether DN should overwrite the downstream
QOP in a write pipeline. This is used in the case where client
talks to first DN with a QOP, but inter-DN communication needs to be
using a different QOP. If set to false, the default behaviour is that
inter-DN communication will use the same QOP as client-DN connection.
dfs.encrypt.data.overwrite.downstream.new.qop
When dfs.datanode.overwrite.downstream.derived.qop is set to true,
this configuration specifies the new QOP to be used to overwrite
inter-DN QOP.
dfs.namenode.blockreport.queue.size
1024
The queue size of BlockReportProcessingThread in BlockManager.
dfs.namenode.storage.dir.perm
700
Permissions for the directories on on the local filesystem where
the DFS namenode stores the fsImage. The permissions can either be
octal or symbolic.
dfs.namenode.blockreport.max.lock.hold.time
4
The BlockReportProcessingThread max write lock hold time in ms.
dfs.namenode.corrupt.block.delete.immediately.enabled
true
Whether the corrupt replicas should be deleted immediately, irrespective
of other replicas on stale storages..
dfs.journalnode.edits.dir.perm
700
Permissions for the directories on on the local filesystem where
the DFS journal node stores the edits. The permissions can either be
octal or symbolic.
dfs.journalnode.handler.count
5
The number of JournalNode RPC server threads that listen to
requests from clients.
dfs.namenode.lease-hard-limit-sec
1200
Determines the namenode automatic lease recovery interval in seconds.
dfs.namenode.gc.time.monitor.enable
true
Enable the GcTimePercentage metrics in NameNode's JvmMetrics. It will
start a thread(GcTimeMonitor) computing the metric.
dfs.namenode.gc.time.monitor.observation.window.ms
1m
Determines the windows size of GcTimeMonitor. A window is a period of time
starts at now-windowSize and ends at now. The GcTimePercentage is the gc
time proportion of the window.
dfs.namenode.gc.time.monitor.sleep.interval.ms
5s
Determines the sleep interval in the window. The GcTimeMonitor wakes up in
the sleep interval periodically to compute the gc time proportion. The
shorter the interval the preciser the GcTimePercentage. The sleep interval
must be shorter than the window size.
dfs.permissions.allow.owner.set.quota
false
Whether the owner(not superuser) of a directory can set quota of his sub
directories when permissions is enabled. Default value is false;
dfs.protected.subdirectories.enable
false
whether to protect the subdirectories of directories which
set on fs.protected.directories.
dfs.storage.default.policy
HOT
Set the default Storage Policy name with following value,
LAZY_PERSIST: memory storage policy.
ALL_SSD : all SSD storage policy.
ONE_SSD : one SSD_storage policy.
HOT : hot storage policy.
WARM : warm policy.
COLD : cold_storage policy.
PROVIDED : provided storage policy.
dfs.datanode.same-disk-tiering.enabled
false
HDFS-15548 to allow DISK/ARCHIVE to be
configured on the same disk mount to manage disk IO.
When this is enabled, datanode will control the capacity
of DISK/ARCHIVE based on reserve-for-archive.percentage.
dfs.datanode.reserve-for-archive.default.percentage
0.0
Default disk capacity ratio of ARCHIVE volume,
expected the value to be between 0 to 1.
This will be applied when DISK/ARCHIVE volumes are configured
on the same mount, which is detected by datanode.
Beware that capacity usage might be >100% if there are already
data blocks exist and the configured ratio is small, which will
prevent the volume from taking new blocks
until capacity is balanced out.
dfs.datanode.same-disk-tiering.capacity-ratio.percentage
Disk capacity ratio of DISK or ARCHIVE volume
when dfs.datanode.same-disk-tiering is turned on
This will override the value of
dfs.datanode.reserve-for-archive.default.percentage .
Example value:
[0.3]/disk1/archive,[0.7]/disk1/disk,[0.4]/disk2/archive,[0.6]/disk2/disk
This is only effective for configured
DISK/ARCHIVE volumes in dfs.datanode.data.dir.
dfs.balancer.getBlocks.hot-time-interval
0
Balancer prefer moving cold blocks i.e blocks associated with files
accessed or modified before the specified time interval.
dfs.datanode.directoryscan.max.notify.count
5
Defines the maximum number of blocks that the DirectoryScanner may notify
namenode right way for received or deleted blocks after one round.
dfs.datanode.nameservices.resolution-enabled
false
Determines if the given nameservice address is a domain name which needs to
be resolved (using the resolver configured by dfs.nameservices.resolver.impl).
This is used by datanode to resolve namenodes.
dfs.datanode.nameservices.resolver.impl
Nameservice resolver implementation used by datanode.
Effective with dfs.nameservices.resolution-enabled on.
dfs.client.mark.slownode.as.badnode.threshold
10
The threshold to mark a slownode as a badnode. If we get PipelineAck from
a slownode continuously for ${dfs.client.treat.slownode.as.badnode.threshold}
times, we should mark it as a badnode.
dfs.datanode.lockmanager.trace
false
If this is true, after shut down datanode lock Manager will print all leak
thread that not release by lock Manager. Only used for test or trace dead lock
problem. In produce default set false, because it's have little performance loss.
dfs.client.fsck.connect.timeout
60000ms
The amount of time the fsck client will wait to connect to the namenode
before timing out.
dfs.client.fsck.read.timeout
60000ms
The amount of time the fsck client will wait to read from the namenode
before timing out. If the namenode does not report progress more
frequently than this time, the client will give up waiting.
dfs.client.output.stream.uniq.default.key
DEFAULT
The default prefix key to construct the uniqKey for one DFSOutputStream.
If the namespace is DEFAULT, it's best to change this conf to other value.
dfs.client.rbf.observer.read.enable
false
Enables observer reads for clients. This should only be enabled when clients are using routers.