- The idea behind NIFI-3356 was to improve the efficiency and throughput of the Provenance Repository, as it is often the bottleneck. While testing the newly designed repository,
a handful of other, fairly minor, changes were made to improve efficiency as well, as these came to light when testing the new repository:
- Use a BufferedOutputStream within StandardProcessSession (via a ClaimCache abstraction) in order to avoid continually writing to FileOutputStream when writing many small FlowFiles
- Updated threading model of MinimalLockingWriteAheadLog - now performs serialization outside of lock and writes to a 'synchronized' OutputStream
- Change minimum scheduling period for components from 30 microseconds to 1 nanosecond. ScheduledExecutor is very inconsistent with timing of task scheduling. With the bored.yield.duration
now present, this value doesn't need to be set to 30 microseconds. This was originally done to avoid processors that had no work from dominating the CPU. However, now that we will yield
when processors have no work, this results in slowing down processors that are able to perform work.
- Allow nifi.properties to specify multiple directories for FlowFile Repository
- If backpressure is engaged while running a batch of sessions, then stop batch processing earlier. This helps FlowFiles to move through the system much more smoothly instead of the
herky-jerky queuing that we previously saw at very high rates of FlowFiles.
- Added NiFi PID to log message when starting nifi. This was simply an update to the log message that provides helpful information.
NIFI-3356: Fixed bug in ContentClaimWriteCache that resulted in data corruption and fixed bug in RepositoryConfiguration that threw exception if cache warm duration was set to empty string
NIFI-3356: Fixed NPE
NIFI-3356: Added debug-level performance monitoring
NIFI-3356: Updates to unit tests that failed after rebasing against master
NIFI-3356: Incorporated PR review feedback
NIFI-3356: Fixed bug where we would delete index directories that are still in use; also added additional debug logging and a simple util class that can be used to textualize provenance event files - useful in debugging
This closes#1493
- Add 'nifi.flow.configuration.archive.max.count' in nifi.properties
- Change default archive limit so that it uses archive max time(30 days)
and storage (500MB) if no limitation is specified
- Simplified logic to delete old archives
This closes#1460.
Signed-off-by: Koji Kawamura <ijokarumawak@apache.org>
Added unit tests for SAN inclusion in CertificateUtils#generateIssuedCertificate() and TlsHelper#generateCertificationRequest().
Fixed typos.
This closes#1491.
Signed-off-by: Andy LoPresto <alopresto@apache.org>
* Remove function based on JDK source.
* Add new function to find bytes based on RFC3629.
* Add field name to log entry when field is truncated.
Signed-off-by: Mike Moser <mosermw@apache.org>
This closes#1475
use the FileNameFilter when not passing down explit jar paths
Filter out ^. files when reading lists of files from directories
Signed-off-by: Koji Kawamura <ijokarumawak@apache.org>
* Updated StandardRecordWriter, even though it is now deprecated to consider the encoding behavior of java.io.DataOutputStream.writeUTF() and truncate string values such that the UTF representation will not be longer than that DataOutputStream's 64K UTF format limit.
* Updated the new SchemaRecordWriter class to similarly truncate long Strings that will be written as UTF.
* Add tests to confirm handling of large UTF strings and various edge conditions of UTF string handling.
Signed-off-by: Mike Moser <mosermw@apache.org>
This closes#1469.
- Support counters at Wait/Notify processors so that NiFi flow can be
configured to wait for N signals
- Extract Wait/Notify logics into WaitNotifyProtocol
- Added FragmentAttributes to manage commonly used fragment attributes
- Changed existing split processors to set 'fragment.identifier' and
'fragment.count', so that Wait can use those to wait for all splits
get processed
This closes#1420.
Signed-off-by: Bryan Bende <bbende@apache.org>
- Removed host and port field from Peer since the same information is
available in PeerDescription
- Refactored variable names in SocketRemoteSiteListener to improve readability
- Changed how SocketRemoteSiteListener constructs PeerDescription
instance. It used to use hard-coded 'localhost' as hostname, and
getPort() which returns server's port. Since the peer is a remote peer,
i.e the client, it should be client hostname and port.
- Added hostname resolution at DataTransferResource to make s2s.host
value consistent with RAW transport. Without this, RAW uses hostname
while HTTP uses IP address. It will be hard to be used from downstream flows.
- Replaced heavy use of mockito which was difficult to maintain, with
nifi-mock
- Added SiteToSiteAttributes and more assertions in unit tests
This closes#1342.
Signed-off-by: Bryan Bende <bbende@apache.org>
NIFI-1582 removing the option to use cluster state
NIFI-1582 addressing Oleg's comments
NIFI-1582 No longer forcing numbers as the init value and adding getStateValue() to EL instead of using 'ATTRIBUTE_state'
NIFI-1582 Removing init state value
NIFI-1582 Adding documentation for the changes to Init State value
This closes#319
- Added urls in addition to the existing url, to support multiple target
URLs
- Backward compatibility is provided by returning the first url if
multipe urls are specified, but component accessing the url doesn't
support multiple urls
- UI is not fully updated yet. Following UI components are planned to be updated
by different commits
- Search component: only the first URL is searchable and shown
- Component status: RPG status shows only the first URL
- Component action history: only the first URL is searchable and shown
- Updated Search component to use URLs.
This closes#1208.
- Changed AtomicBoolean to CountDownLatch to avoid sleeping thread in
some test cases
- Specified less number of threads for Jetty and LittleProxy than
default to lower resource usage
- Added try catch for the specific gateway timeout case (504) so that
test can pass even it happens while it fails with other errors
NIFI-2854: Incorporated PR review feedback
NIFI-2854: Implemented feedback from PR Review
NIFI-2854: Ensure that all resources are closed on CompressableRecordReader.close() even if an IOException is thrown when closing one of them
This closes#1202
- introduced org.apache.nifi.stream.io.util.TextLineDemarcator
- refactored SplitText to use org.apache.nifi.stream.io.util.TextLineDemarcator
- updated SplitText's capability discription to provide more clarity around splits with headers.
NIFI-1712 Applying per-instance class loading to HBaseClientService to allow specifying Phoenix Client JAR
-Refactoring the ClassLoading so that every processor, controller service, and reporting task gets an InstanceClassLoader with a parent of the NAR ClassLoader, and only components with @RequiresInstanceClassLoading will make a copy of the NAR ClassLoader resources, and addressing some review feedback
This closes#1156
* Java version and vendor
* OS name and version
* Release Tag
* Build revision (commit SHA), branch, and timestamp
* Handles formal releases, ad-hoc builds, and non-release source builds
* Standalone UI presence in About dialog, Summary -> System Diagnostics
* Cluster UI as Versions tab in Cluster dialog
* Reduce About Dialog Content
* Fix Missing Property Display Bugs
* Marking the build time as type string.
* This closes#583
- Consolidated the target cluster URL resolving logic into
SiteToSiteRestApiClient's as a common method
- Changed to more descriptive error message
- Added more unit test cases
- Added following properties:
- nifi.web.http.port.forwarding
- nifi.web.https.port.forwarding
This closes#1100.
Signed-off-by: Koji Kawamura <ijokarumawak@apache.org>
This commit fixes following two issues, that happens when a Root Group Port
policy for S2S data transfer is removed at a remote NiFi, after a client NiFi has
connected to that port:
1. At client side, Remote Process Group should show that authorization
is failing on its bulletin, but the Exception is caught and
ignored. Nothing is shown on the UI with HTTP transport protocol.
RAW S2S shows error on RPG bulletin. This commit fixes HTTP S2S to
behave the same.
2. At server side, corresponding input-port or output-port should show
that it is accessed by an unauthorized client on its bulletin, but it's
not shown with HTTP transport protocol.
RAW S2S shows warning messages for this. This commit fixes HTTP S2S to
behave the same.
In order to fix the 2nd issue above, request authorization at
DataTransferResource is changed from using DataTransferAuthorizable
directly, to call RootGroupPort.checkUserAuthorization().
Because the blettin is tied to the Port instance and it's
difficult to produce blettin message from this resource.
Since RootGroupPort.checkUserAuthorization uses
DataTransferAuthorizable inside, the check logic stays the same as
before.
Adding a RootGroupPortAuthorizable to provide access to necessary components for performing the authorization.
This closes#996
- Fixing contrib check issues.
- Clean up pom.
- Addressing issue where reporting task property descriptor using wrong scope.
NIFI-2635:
- Fixing issue with revisions when creating users and user groups.
- Forwarding requests to the coordinator instead of replicating.
- Tweaking verbage in dialog for removing users and groups.
This closes#943
Without this fix, NiFi fails to send data via HTTP Site-to-Site through
Proxy which requires authentication due to AsynchronousCloseException.
It happens when async client replays producing contents in order to re-send the
request with auth credential for the proxy server, however the
connection is already closed.
This fix makes NiFi to send actual data only at the second round of requests, so that flow-file
contents can be sent without reading it twice.
Signed-off-by: Yolanda M. Davis <ymdavis@apache.org>
This closes#915
Refresh remote peer statuses even if the bootstrap node goes down.
Migrate existing code which handles the situation from
EndpointConnectionPool to PeerSelector, so that both RAW and HTTP
transport protocol has the same capability.
This closes#927.
- It couldn't send data larger than about 7KB due to the mis-use of
httpasyncclient library
- Updated httpasyncclient from 4.1.1 to 4.1.2
- Let httpasyncclient framework to call produceContent multiple times as
it gets ready to send more data via SSL session
- Added HTTPS test cases to TestHttpClient, which failed without this
fix
- Logic for sorting DN, reversing X500Names before using them to generate certificate
- Logging reordered dn
- Accounting for limited crypto pkcs12, allowing password specification for client certificates
- Updating tests to work with or without jce unlimited
- Loading keystore for test in try-with
This closes#824.
Signed-off-by: Bryan Bende <bbende@apache.org>
- Removed message queueing which could result in data loss
- Fixed life-cycle issues that coudl put processor in an unstable state
- Fixed PropertyDescriptor translation for Time units and Byte sizes
- Fixed broken tests
- Added additional tests
NIFI-2519 added default for SMTP_MAXIMUM_CONNECTIONS
NIFI-2519 addressed PR comments, polishing
- fixed intermittent deadlock on processor stop and added test for it
- the attributes that can not be extracted from the message but available via MessageContext are written into the outgoing FlowFile
- other minor fixes
NIFI-2519 addressed lates PR comments
NIFI-2519 added better messaging when server closes the connection
NIFI-2519 some polishing and additional tests to validate deadlocks
NIFI-2519 address latest PR comments
fixed deadlock condition for when the consumer is stopped while server is distributing messages
fixed MAX message size issue ensuring it is validated
set max connections to SMTPServer
polished pom
added L&N
NIFI-2519 PR comments
- fixed LICENSE
- Added usage of LimitingInputStream
- simplified SmtpConsumer by removing hasMessage operation
This closes#695.
Signed-off-by: Andy LoPresto <alopresto@apache.org>
Defaulting to same keyStore, key password (+18 squashed commits)
Squashed commits:
[9d01ba0] NIFI-2193 - Fixing typo
[55440bc] NIFI-2193 - Standalone can run as long as there are no conflicting files/folders
[0ca34ed] NIFI-2193 - Fixing some filename, absolute path issues
[9d4f65b] NIFI-2193 - Incorporating feedback
[f7550b4] NIFI-2193 - Cleaning up imports
[59a7637] NIFI-2193 - Updating umask to allow owner to execute
[cf824e7] NIFI-2193 - Moving DN arg to CA service specific parent class
[921ee13] NIFI-2193 - Making keystore getInstance more consistent
[a283c4b] NIFI-2193 - Updating sample config files in assembly to reflect new structure
[8d3a21d] NIFI-2193 - Making TlsHelper static, adding option to use same password for Key, KeyStore
[b13d247] NIFI-2193 - Addressing PR feedback
[46ef8ed] NIFI-2193 - Removing commons-logging, log4j from notice
[d4cf41a] NIFI-2193 - Adding option to specify output file for CA certificate when using cli client
[b74bf25] NIFI-2193 - Removing Bouncy Castle from notice
[6e34f9a] NIFI-2193 - Adding CLI client for easier generation of client certificates
[2924fca] NIFI-2193 - nifi-toolkit-ssl -> nifi-toolkit-tls, removing unused constants
[886167e] NIFI-2193 - Adding slf4j to avoid runtime issue
[082de46] NIFI-2193 - Command line SSL config utility as well as certificate authority client/server
Fixed NPE (+2 squashed commits)
Squashed commits:
[c5d521a] NIFI-2466: Added unit test to verify changes; fixed validation
[aa4d418] NIFI-2446: Add option to specify key password when different than keystore password
This closes#776.
Signed-off-by: Andy LoPresto <alopresto@apache.org>
Fixed Site-to-Site Transit URI for HTTP to be consistent with RAW socket.
- Removed url from CommunicationsSession since it's redundant as we have
Peer.url, too. The value was not used from anywhere other than HTTP
Site-to-Site.
- Added createTransitUri method in Communicant interface, so that
implementation can customize transitUri while providing consistent
interface.
- added SnippetUtilsTest
- renamed TypeOneUUIDGenerator to ComponentIdGenerator
- changed lsb part of ComponentIdGenerator back to long
- Fixed 'isCopy' condition for clustered environments
This closes#718.