If a node exceeds the flood-stage disk watermark then we add a block to all of
its indices to prevent further writes as a last-ditch attempt to prevent the
node completely exhausting its disk space. However today this block remains in
place until manually removed, and this block is a source of confusion for users
who current have ample disk space and did not even realise they nearly ran out
at some point in the past.
This commit changes our behaviour to automatically remove this block when a
node drops below the high watermark again. The expectation is that the high
watermark is some distance below the flood-stage watermark and therefore the
disk space problem is truly resolved.
Fixes#39334
Previously persistent task assignment was checked in the
following situations:
- Persistent tasks are changed
- A node joins or leaves the cluster
- The routing table is changed
- Custom metadata in the cluster state is changed
- A new master node is elected
However, there could be situations when a persistent
task that could not be assigned to a node could become
assignable due to some other change, such as memory
usage on the nodes.
This change adds a timed recheck of persistent task
assignment to account for such situations. The timer
is suspended while checks triggered by cluster state
changes are in-flight to avoid adding burden to an
already busy cluster.
Closes#35792
The new limit on the number of open shards in a cluster may be
interpreted by users as a sizing recommendation, but it is not. This
clarifies in the documentation that this is a safety limit, not a
recommendation.
This removes the option to run a cluster without enforcing the
cluster-wide shard limit, making strict enforcement the default and only
behavior. The limit can still be adjusted as desired using the cluster
settings API.
In a future major version, we will be introducing a soft limit on the
number of shards in a cluster based on the number of nodes in the
cluster. This limit will be configurable, and checked on operations
which create or open shards and issue a warning if the operation would
take the cluster over the limit.
There is an option to enable strict enforcement of the limit, which
turns the warnings into errors. In a future release, the option will be
removed and strict enforcement will be the default (and only) behavior.
As user-defined cluster metadata is accessible to anyone with access to
get the cluster settings, stored in the logs, and likely to be tracked
by monitoring solutions, it is useful to clarify in the documentation
that it should not be used to store secret information.
Adds a place for users to store cluster-wide data they wish to associate
with the cluster via the Cluster Settings API. This is strictly for
user-defined data, Elasticsearch makes no other other use of these
settings.
Control max size and count of warning headers
Add a static persistent cluster level setting
"http.max_warning_header_count" to control the maximum number of
warning headers in client HTTP responses.
Defaults to unbounded.
Add a static persistent cluster level setting
"http.max_warning_header_size" to control the maximum total size of
warning headers in client HTTP responses.
Defaults to unbounded.
With every warning header that exceeds these limits,
a message will be logged in the main ES log,
and any more warning headers for this response will be
ignored.
This commit adds a new setting `cluster.persistent_tasks.allocation.enable`
that can be used to enable or disable the allocation of persistent tasks.
The setting accepts the values `all` (default) or `none`. When set to
none, the persistent tasks that are created (or that must be reassigned)
won't be assigned to a node but will reside in the cluster state with
a no "executor node" and a reason describing why it is not assigned:
```
"assignment" : {
"executor_node" : null,
"explanation" : "persistent task [foo/bar] cannot be assigned [no
persistent task assignments are allowed due to cluster settings]"
}
```
Update allocation awareness docs
Today, the docs imply that if multiple attributes are specified the the
whole combination of values is considered as a single entity when
performing allocation. In fact, each attribute is considered separately. This
change fixes this discrepancy.
It also replaces the use of the term "awareness zone" with "zone or domain", and
reformats some paragraphs to the right width.
Fixes#29105
Cluster settings shouldn't leak into the next test.
I played with failing the test if it left over any settings but that
felt like it added more ceremony then it was worth. The advantage is
that any test that intentionally wants to leave settings in place after
the test would fail and require looking at but, so far as I can tell, we
don't have any such tests.
Clear the disk watermark after the snippet showing users how to set it.
Without this our tests will fail if the disks have less than 10GB free.
Closes#28325
This commit adds cross-settings validation for the low/high/flood stage
disk watermark settings. This validation was enabled by the introduction
of multiple settings validation.
Relates #25600
Today when we run out of disk all kinds of crazy things can happen
and nodes are becoming hard to maintain once out of disk is hit.
While we try to move shards away if we hit watermarks this might not
be possible in many situations. Based on the discussion in #24299
this change monitors disk utilization and adds a flood-stage watermark
that causes all indices that are allocated on a node hitting the flood-stage
mark to be switched read-only (with the option to be deleted). This allows users to react on the low disk
situation while subsequent write requests will be rejected. Users can switch
individual indices read-write once the situation is sorted out. There is no
automatic read-write switch once the node has enough space. This requires
user interaction.
The flood-stage watermark is set to `95%` utilization by default.
Closes#24299
Today when an index is `read-only` the index is also blocked from
being deleted which sometimes is undesired since in-order to make
changes to a cluster indices must be deleted to free up space. This is
a likely scenario in a hosted environment when disk-space is limited to switch
indices read-only but allow deletions to free up space.
* plugins/discovery-azure-class.asciidoc
* reference/cluster.asciidoc
* reference/modules/cluster/misc.asciidoc
* reference/modules/indices/request_cache.asciidoc
After this is merged there will be no unconvereted snippets outside
of `reference`.
Related to #18160
`cluster.routing.allocation.cluster_concurrent_rebalance` setting,
clarifying in which shard allocation situations the rebalance limit
takes effect.
Closes#20529
In 7560101ec7, the Elasticsearch logger
names were modified to be their fully-qualified class name (with some
exceptions for special loggers like the slow logs and the transport
tracer). This commit updates the docs accordingly.
Relates #20475
Today when parsing settings during bootstrap, we add a system property
for every Elasticsearch setting. Additionally, settings can be set via
system properties. This commit simplifies this situation.
- settings are no longer propogated to system properties
- system properties can not be used to set settings
- the "es." prefix on settings is no longer required (nor permitted)
- test logging has a dedicated system property (tests.logger.level)
Relates #18198
Adds infrastructure so `gradle :docs:check` will extract tests from
snippets in the documentation and execute the tests. This is included
in `gradle check` so it should happen on CI and during a normal build.
By default each `// AUTOSENSE` snippet creates a unique REST test. These
tests are executed in a random order and the cluster is wiped between
each one. If multiple snippets chain together into a test you can annotate
all snippets after the first with `// TEST[continued]` to have the
generated tests for both snippets joined.
Snippets marked as `// TESTRESPONSE` are checked against the response
of the last action.
See docs/README.asciidoc for lots more.
Closes#12583. That issue is about catching bugs in the docs during build.
This catches *some* bugs in the docs during build which is a good start.
Today, certain bootstrap properties are set and read via system
properties. This action-at-distance way of managing these properties is
rather confusing, and completely unnecessary. But another problem exists
with setting these as system properties. Namely, these system properties
are interpreted as Elasticsearch settings, not all of which are
registered. This leads to Elasticsearch failing to startup if any of
these special properties are set. Instead, these properties should be
kept as local as possible, and passed around as method parameters where
needed. This eliminates the action-at-distance way of handling these
properties, and eliminates the need to register these non-setting
properties. This commit does exactly that.
Additionally, today we use the "-D" command line flag to set the
properties, but this is confusing because "-D" is a special flag to the
JVM for setting system properties. This creates confusion because some
"-D" properties should be passed via arguments to the JVM (so via
ES_JAVA_OPTS), and some should be passed as arguments to
Elasticsearch. This commit changes the "-D" flag for Elasticsearch
settings to "-E".
Today we throttle recoveries only for incoming recoveries. Nodes that have a lot
of primaries can get overloaded due to too many recoveries. To still keep that at bay
we limit the number of threads that are sending files to the target to overcome this problem.
The right solution here is to also throttle the outgoing recoveries that are today unbounded on
the master and don't start the recovery until we have enough resources on both source and target nodes.
The concurrency aspects of the recovery source also added a lot of complexity and additional threadpools
that are hard to configure. This commit removes the concurrent streamns notion completely and sends files
in the thread that drives the recovery simplifying the recovery code considerably.
Outgoing recoveries are not throttled on the master via a allocation decider.
Prior to 2.0 we summed up the available space on all disk on a node
due to the raid-0 like behavior. Now we don't do this anymore and use the
min & max disk space to make decisions.
Closes#13106
Bukhtawar
David Roberts
debadair
Gordon Brown
David Turner
Mayya Sharipova
Tanguy Leroux
Nik Everett
Glen Smith
Aron Szanto
Clinton Gormley
Jason Tedor
Herman Schaaf
Simon Willnauer
Boaz Leskes
Ali Beyad
Joshua Rich
Brandon Wulf
Dongjoon Hyun
Yannick Welsch
Masaru Hasegawa