controller.git
9 years agoBug 865: Fixed use of removed deprecated YANGInstanceIdentifier methods. 47/12047/4
Tony Tkacik [Fri, 17 Oct 2014 12:05:06 +0000 (14:05 +0200)]
Bug 865: Fixed use of removed deprecated YANGInstanceIdentifier methods.

Some components in controller still used methods which was deprecated
before Helium and was removed after Helium was released to accomodate
API changes for 0.7.0 Yangtools. This patch migrates code to replacement
methods.

Change-Id: I146a08caefeb0d666223cf66600a4342a5b49165
Signed-off-by: Tony Tkacik <ttkacik@cisco.com>
9 years agoMerge "Migrate netconf testtool to work with karaf distribution"
Tony Tkacik [Fri, 17 Oct 2014 13:00:50 +0000 (13:00 +0000)]
Merge "Migrate netconf testtool to work with karaf distribution"

9 years agoMerge "Bug 1764 - added service for pushing default node configuration"
Tony Tkacik [Fri, 17 Oct 2014 10:56:13 +0000 (10:56 +0000)]
Merge "Bug 1764 - added service for pushing default node configuration"

9 years agoMerge "Add features-netconf-connector to controller karaf distro"
Tony Tkacik [Fri, 17 Oct 2014 10:13:30 +0000 (10:13 +0000)]
Merge "Add features-netconf-connector to controller karaf distro"

9 years agoMerge "BUG-2006: Rework state tracking"
Tony Tkacik [Fri, 17 Oct 2014 09:17:35 +0000 (09:17 +0000)]
Merge "BUG-2006: Rework state tracking"

9 years agoMigrate netconf testtool to work with karaf distribution 32/11532/4
Maros Marsalek [Thu, 16 Oct 2014 15:41:07 +0000 (17:41 +0200)]
Migrate netconf testtool to work with karaf distribution

Make testtool part of default build process.

Change-Id: Ie957bcf2a38405fee979a6f8547b328d3bd8e2ea
Signed-off-by: Maros Marsalek <mmarsale@cisco.com>
9 years agoAdd features-netconf-connector to controller karaf distro 27/12027/2
Maros Marsalek [Thu, 16 Oct 2014 15:40:37 +0000 (17:40 +0200)]
Add features-netconf-connector to controller karaf distro

Change-Id: I44bf8d82f35586a03950d5ef2f26a321931cb8c6
Signed-off-by: Maros Marsalek <mmarsale@cisco.com>
9 years agoBug 2038: Ensure only one concurrent 3-phase commit in Shard 95/11795/10
tpantelis [Sun, 5 Oct 2014 03:05:16 +0000 (23:05 -0400)]
Bug 2038: Ensure only one concurrent 3-phase commit in Shard

Added a ShardCommitCoordinator class that ensures there's only one
concurrent 3-phase commit.

The following outlines the new commit workflow:

- On ready, the ShardTransaction creates the dom store cohort and
forwards a new ForwardedReadyTransaction message to the shard.

- The shard calls its ShardCommitCoordinator to add the cohort and
modificaton to a cached keyed by transaction ID.

- On CanCommitTransaction message, the ShardCommitCoordinator looks up and removes
the cohort entry from the cache corresponding to the transaction ID
passed via the CanCommit message. The ShardCommitCoordinator also caches
the cohort entry for the current transaction in progress. If there's no
transaction in progress, the committing transaction becomes the current
transaction and canCommit is called on the cohort. Otherwise, the cohort
entry is queued to be processed after the current tranaction completes.

- On CommitTransaction message, if the transaction ID passed via the
Commit message matches the currently cached cohort entry, the preCommit
and commit phases are performed. When complete, the ShardCommitCoordinator
dequeues the next waiting transaction cohort entry, if any, and process
it.

If a Tx is aborted and it is the current transaction, the
ShardCommitCoordinator handles it as a completed Tx.

Implemented a timeout mechanism using the akka scheduler such that if
the commit message isn't received after a period of time (default 30 s)
after the canCommit message, the transaction is aborted so that the next
transaction can proceed. This is to handle remote node or network
failures during a 3-phrase commit.

The ThreePhaseCommitCohort actor was removed along with the
ForwardedCommitTransaction.

Change-Id: Iaa5692ca45cd7635d1a06a609f4bf98bec50df14
Signed-off-by: tpantelis <tpanteli@brocade.com>
9 years agoMerge "Bug 1965: Fixed DataChangedReply sent to deadletters"
Moiz Raja [Thu, 16 Oct 2014 20:51:59 +0000 (20:51 +0000)]
Merge "Bug 1965: Fixed DataChangedReply sent to deadletters"

9 years agoMerge "Bug-1607: Clustering : Remove actorFor (deprecated) call from TransactionProx...
Moiz Raja [Thu, 16 Oct 2014 17:03:23 +0000 (17:03 +0000)]
Merge "Bug-1607:  Clustering : Remove actorFor (deprecated) call from TransactionProxy.java"

9 years agoBug 1764 - added service for pushing default node configuration 00/12000/4
Martin Bobak [Thu, 16 Oct 2014 05:29:05 +0000 (07:29 +0200)]
Bug 1764 - added service for pushing default node configuration

Change-Id: I8018c639ce05118722c8f33ca842436cca44372e
Signed-off-by: Martin Bobak <mbobak@cisco.com>
9 years agoMerge "BUG 2047 - HTTP GET - no returning error message"
Tony Tkacik [Thu, 16 Oct 2014 13:21:40 +0000 (13:21 +0000)]
Merge "BUG 2047 - HTTP GET - no returning error message"

9 years agoMerge "BUG 1966 - change message logging level (info -> trace)"
Tony Tkacik [Thu, 16 Oct 2014 13:03:16 +0000 (13:03 +0000)]
Merge "BUG 1966 - change message logging level (info -> trace)"

9 years agoMerge "Bug 2131 - NPE when starting controller"
Tony Tkacik [Thu, 16 Oct 2014 13:02:09 +0000 (13:02 +0000)]
Merge "Bug 2131 - NPE when starting controller"

9 years agoBUG-2006: Rework state tracking 17/11417/5
Robert Varga [Thu, 18 Sep 2014 13:53:24 +0000 (15:53 +0200)]
BUG-2006: Rework state tracking

Separate out the two embedded classes to allow easier analysis of
interactions. Introduce a state-tracking class within
DOMStoreTransactionChainImpl with three distinct states: Idle, Allocated
and Shutdown. The allocated state has an internal sub-state when the
transaction has been readied. Perform atomic state transitions using
compare-and-swap.

Change-Id: I4b852d42455d7e07076c56af3626f804ba572770
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoMerge changes I879118ce,I664b391e
Tony Tkacik [Thu, 16 Oct 2014 10:09:05 +0000 (10:09 +0000)]
Merge changes I879118ce,I664b391e

* changes:
  BUG-2017 Fix possible concurrent modification ex in config subsystem
  BUG-1520 Set correct factory in ImmediateEventExecutorModuleTest

9 years agoMerge "BUG-1958 Fix leafref path in config-yang module."
Tony Tkacik [Thu, 16 Oct 2014 10:08:50 +0000 (10:08 +0000)]
Merge "BUG-1958 Fix leafref path in config-yang module."

9 years agoMerge "BUG-1521 netconf-impl & netconf-monitoring line coverage."
Tony Tkacik [Thu, 16 Oct 2014 10:08:13 +0000 (10:08 +0000)]
Merge "BUG-1521 netconf-impl & netconf-monitoring line coverage."

9 years agoBug-1607: Clustering : Remove actorFor (deprecated) call from TransactionProxy.java 34/11834/4
Kamal Rameshan [Wed, 8 Oct 2014 21:09:37 +0000 (14:09 -0700)]
Bug-1607:  Clustering : Remove actorFor (deprecated) call from TransactionProxy.java

1. Changes to use ActorSelection instead of ActorPath
2. Removed actorFor usage
3. Removed resolvePath calls
4. Changes in test classes.

Rebased.

Change-Id: I8fdc18a1ac18d75d1b6d5c8cd0da986e19d08280
Signed-off-by: Kamal Rameshan <kramesha@cisco.com>
9 years agoUpdate the XSQL jdbc driver to support pentaho 24/11724/5
Sharon Aicler [Fri, 3 Oct 2014 15:36:10 +0000 (08:36 -0700)]
Update the XSQL jdbc driver to support pentaho

Change-Id: I72f54b36a87c84c9fa493a1d35300b241a53ec34
Signed-off-by: Sharon Aicler <saichler@cisco.com>
9 years agoBug 1965: Fixed DataChangedReply sent to deadletters 32/11932/3
tpantelis [Mon, 6 Oct 2014 22:31:18 +0000 (18:31 -0400)]
Bug 1965: Fixed DataChangedReply sent to deadletters

Change-Id: I3a579823d57e113597aabeb647c883d8811f29f1
Signed-off-by: tpantelis <tpanteli@brocade.com>
9 years agoBug 2160: Added concurrent 3-phase commit coordinator 66/11966/3
tpantelis [Tue, 7 Oct 2014 13:34:01 +0000 (09:34 -0400)]
Bug 2160: Added concurrent 3-phase commit coordinator

Added a new class DOMConcurrentDataCommitCoordinator which is an
implementation of DOMDataCommitExecutor that allows multiple concurrent
3-phase transaction commits. The 3 phases are still performed serially and
non-blocking per transaction.

A new flag, "allow-concurrent-commits", was added to the DOM in-memory
data broker's yang config. If true, the new
DOMConcurrentDataCommitCoordinator implementation is used, otherwise the
current implementation is used. The default is false.

The distributed data store will use the concurrent coordinator.

Change-Id: Ie21947d6f9c1086f89686a73442c854c08c07100
Signed-off-by: tpantelis <tpanteli@brocade.com>
9 years agoIncrement the AAA dependency by 0.1.0 for post-Helium master branch 66/11766/2
Colin Dixon [Tue, 7 Oct 2014 02:57:14 +0000 (21:57 -0500)]
Increment the AAA dependency by 0.1.0 for post-Helium master branch

This patch is the second half of the post-Helium version bump patch. It
does two things:
* Increments the version of the depenency on AAA from 0.1.0-SNAPSHOT to
  0.2.0-SNAPSHOT.
* Re-enables the feature tests for the restconf and netconf-conector
  features.

Change-Id: I3160e429a4368ee142a41deb23631f6c1455d358
Signed-off-by: Colin Dixon <colin@colindixon.com>
9 years agoIncrementing versions by 0.1.0 for post-Helium master branch 39/11739/4
Colin Dixon [Fri, 3 Oct 2014 22:55:46 +0000 (17:55 -0500)]
Incrementing versions by 0.1.0 for post-Helium master branch

* The only version not incrmented is aaa.version, which is left at
  0.1.0-SNAPSHOT since they depend on controller and can't update yet.
* To break the cyclic dependency, this patch temporarily stops running
  the netconf-connector and restconf feature tests.
* Also fixing a bug causing Errors like "Choice has two nodes case with
  same qnames".

A second patch (to be run after AAA increments their versions to
0.2.0-SNAPSHOT) wiil update aaa.version and re-enable these tests.

Change-Id: I70a3987d32504e798c5ed05a67a0a1174338844b
Signed-off-by: Colin Dixon <colin@colindixon.com>
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoBug 2131 - NPE when starting controller 09/11709/2
Jan Hajnar [Wed, 1 Oct 2014 14:49:25 +0000 (16:49 +0200)]
Bug 2131 - NPE when starting controller

* made InvenroryAndReadAdapter wait for BindingAwareBroker because it
was calling DeteBrokerService before it was available

Change-Id: Ie38415194544997bc12d2dd3acfc57bfa8918730
Signed-off-by: Jan Hajnar <jhajnar@cisco.com>
9 years agoBug 2107: Use DataCommitListeners for inventory events 05/11705/5
Ed Warnicke [Wed, 1 Oct 2014 16:40:29 +0000 (09:40 -0700)]
Bug 2107: Use DataCommitListeners for inventory events

Change-Id: I648bba5e7585cc9d1a70f3dcaf6f8fdbcfeb1787
Signed-off-by: Ed Warnicke <eaw@cisco.com>
9 years agoFixes bug 2114 88/11688/1
Abhishek Kumar [Wed, 1 Oct 2014 05:16:08 +0000 (22:16 -0700)]
Fixes bug 2114

Adds handling of "leaf" node at the module level.

Change-Id: I55efb47b57d2f33b136b0dae8c82effe04ad706b
Signed-off-by: Abhishek Kumar <abhishk2@cisco.com>
9 years agoMerge changes I95a821aa,I4a90a3b6,I9f77a16a
Tony Tkacik [Tue, 30 Sep 2014 17:17:39 +0000 (17:17 +0000)]
Merge changes I95a821aa,I4a90a3b6,I9f77a16a

* changes:
  BUG-2078  Stats not being collected from all nodes, dangling nodes left in oper data store after mininet disconnects BUG-2049  DataStore failure in StatisticsManager
  Bug 2106: Refactored FlowCapableInventoryProvider to properly work with failures.
  Bug 2103: Revert "Bug 2101: Do not block processing on transaction submit."

9 years agoBUG-2109 : removed final modifier to allow overriding of negotiationFailed() method 69/11669/1
Dana Kutenicsova [Tue, 30 Sep 2014 13:16:21 +0000 (15:16 +0200)]
BUG-2109 : removed final modifier to allow overriding of negotiationFailed() method

Change-Id: Iffe9dcf061480844ceb8a8278997c28000ec329d
Signed-off-by: Dana Kutenicsova <dkutenic@cisco.com>
9 years agoBUG-2078 Stats not being collected from all nodes, dangling nodes left in oper data... 64/11664/4
Vaclav Demcak [Mon, 29 Sep 2014 20:47:24 +0000 (22:47 +0200)]
BUG-2078  Stats not being collected from all nodes, dangling nodes left in oper data store after mininet disconnects
BUG-2049  DataStore failure in StatisticsManager

* fix performance issues
* fix occurrence zombie nodes
* fix NPE for StatCollector
* fix NPE for Meter/Group lists

Change-Id: I95a821aaf308bdb6e989c4a740aa014d5aaab4c2
Signed-off-by: Vaclav Demcak <vdemcak@cisco.com>
9 years agoBug 2106: Refactored FlowCapableInventoryProvider to properly work with failures. 63/11663/2
Tony Tkacik [Mon, 29 Sep 2014 17:04:40 +0000 (19:04 +0200)]
Bug 2106: Refactored FlowCapableInventoryProvider to properly work with failures.

FlowCapableInventoryProvider was unable to continue if transaction chain
failed during construction of new transaction.

Refactored it in way, that it is able to reconstruct and retry transaction,
which failed because of previous transaction and retries it once
in new transaction chain. If it fails again, it continues with next batch.

Change-Id: I4a90a3b60b0a49d562c8648a779e790f55840b5d
Signed-off-by: Tony Tkacik <ttkacik@cisco.com>
9 years agoBug 2103: Revert "Bug 2101: Do not block processing on transaction submit." 59/11659/1
Ed Warnicke [Mon, 29 Sep 2014 07:06:35 +0000 (00:06 -0700)]
Bug 2103: Revert "Bug 2101: Do not block processing on transaction submit."

Turns out that the get was throttling the stats manager so
it didn't overwhelm the queue.

This reverts commit 6b3c2a9e446bf91942a48427e5592909a173a399.

Change-Id: I9f77a16a959594c65ed3e61b1265c56da620c1ba
Signed-off-by: Ed Warnicke <eaw@cisco.com>
9 years agoMerge "Bug 2103: Make sure InventoryManager survives Transaction failure."
Ed Warnicke [Sat, 27 Sep 2014 23:09:49 +0000 (23:09 +0000)]
Merge "Bug 2103: Make sure InventoryManager survives Transaction failure."

9 years agoMerge "Bug 2101: Do not block processing on transaction submit."
Ed Warnicke [Sat, 27 Sep 2014 22:47:17 +0000 (22:47 +0000)]
Merge "Bug 2101: Do not block processing on transaction submit."

9 years agoMerge "BUG 2049 DataStore failure in StatisticsManager BUG 2101 When L2 switch...
Ed Warnicke [Sat, 27 Sep 2014 22:46:50 +0000 (22:46 +0000)]
Merge "BUG 2049  DataStore failure in StatisticsManager BUG 2101  When L2 switch installed, its flows are interpreted as new flows with each stats request"

9 years agoBug 2103: Make sure InventoryManager survives Transaction failure. 28/11628/1
Tony Tkacik [Sat, 27 Sep 2014 16:13:21 +0000 (18:13 +0200)]
Bug 2103: Make sure InventoryManager survives Transaction failure.

Change-Id: I5896a0b7c3e63d2828361dc93d3582843a35b35d
Signed-off-by: Tony Tkacik <ttkacik@cisco.com>
9 years agoBug 2101: Do not block processing on transaction submit. 27/11627/3
Tony Tkacik [Sat, 27 Sep 2014 15:25:23 +0000 (17:25 +0200)]
Bug 2101: Do not block processing on transaction submit.

Change-Id: Ibdc75aa670254f84aa70647d655e4e1438edd9e9
Signed-off-by: Tony Tkacik <ttkacik@cisco.com>
9 years agoBUG 2049 DataStore failure in StatisticsManager 98/11598/13
Vaclav Demcak [Thu, 25 Sep 2014 21:03:30 +0000 (23:03 +0200)]
BUG 2049  DataStore failure in StatisticsManager
BUG 2101  When L2 switch installed, its flows are interpreted as new flows with each stats request

* hashCode is not safe in general so we'd like to prevent Collisions how we can
* we change HashBuilding to KeyBuilding from toString method
* toString is generated in CompileTime
* it can not be change during JVM run
* every one use same YANG model version and jar files
* and the same aproach could by use across claster
* add cleaning disconnected Node from StatListeningCommiter implementers

Change-Id: I645e9f07382af0b293bc43698446b7b84b95bbde
Signed-off-by: Vaclav Demcak <vdemcak@cisco.com>
9 years agoBUG 2074 : clustering.services-implementation failed to register listner. 21/11621/1
Moiz Raja [Sat, 27 Sep 2014 00:53:10 +0000 (17:53 -0700)]
BUG 2074 : clustering.services-implementation failed to register listner.

The adsal feature depends on both the stub and implementaton of clustering.
This causes a problem when deployed because sometimes the stub is wired in first
which leads to createCache requests going to that implementation. Subsequently
the real implementation gets wired in and add listener calls go there but since
the caches were created on the stub this fails.

This patch removes the stub as a dependency with the assumption that it is not
required in production.

Change-Id: I42203bbf31b46560c23f283ddd83e56bd950830b
Signed-off-by: Moiz Raja <moraja@cisco.com>
9 years agoBug 1577: Gates access to Shard actor until its initialized 15/11415/4
Abhishek Kumar [Sat, 20 Sep 2014 08:19:11 +0000 (01:19 -0700)]
Bug 1577: Gates access to Shard actor until its initialized

1. Shard manager creates Shard and mark them un-initialized.
   Shard completes recovery and onRecoveryComplete, sends a
   message to Shard manager to mark it initialized.
   If a request for Shard comes to Shard manager and the
   shard is not initialized, it sends ActorNotInitialized
   message.

2. Normalizes and refactors ActorContext.

3. Adds AbstractUntypedPersistentActorWithMetering to meter
   ShardManager.

Change-Id: Ibf15a2ef56422bda53067039d2271a719b6b2ce3
Signed-off-by: Abhishek Kumar <abhishk2@cisco.com>
9 years agoBUG-1958 Fix leafref path in config-yang module. 88/11588/1
Maros Marsalek [Thu, 25 Sep 2014 15:29:57 +0000 (17:29 +0200)]
BUG-1958 Fix leafref path in config-yang module.

When using a condition in the path, the value that is compared has to be a leaf reaf

Change-Id: I898b9448099fd7240aea49cc1c3fe40e9ee7c3d0
Signed-off-by: Maros Marsalek <mmarsale@cisco.com>
9 years agoBUG 2047 - HTTP GET - no returning error message 63/11563/1
Jozef Gloncak [Thu, 25 Sep 2014 06:19:36 +0000 (08:19 +0200)]
BUG 2047 - HTTP GET - no returning error message

Error message was added to NormalizedNodeXmlBodyWriter.

Change-Id: I43914eaa62149b66303cef956f73e694fae9217f
Signed-off-by: Jozef Gloncak <jgloncak@cisco.com>
9 years agoBUG-2049: DataStore failure in StatisticsManager 39/11539/2
Michal Rehak [Wed, 24 Sep 2014 16:21:38 +0000 (18:21 +0200)]
BUG-2049: DataStore failure in StatisticsManager

!WORKAROUND ONLY

- during node reconciliation statsManager could join together 2 similar
  sets of statistics notifications and hit duplicate values during
  processing - here we just skip those duplicates in order not to loose
  stats working thread and finish actual multipart

Change-Id: I526ab5439055ac565bfaa08e84e5043e8764846a
Signed-off-by: Michal Rehak <mirehak@cisco.com>
9 years agoBUG 1966 - change message logging level (info -> trace) 22/11522/1
Jozef Gloncak [Wed, 24 Sep 2014 13:11:31 +0000 (15:11 +0200)]
BUG 1966 - change message logging level (info -> trace)

Change-Id: Idff363361f5d283edf25ac656a519bc8232a9393
Signed-off-by: Jozef Gloncak <jgloncak@cisco.com>
9 years agoMerge "Bug 1948: Finish separating out restconf features"
Moiz Raja [Wed, 24 Sep 2014 05:20:26 +0000 (05:20 +0000)]
Merge "Bug 1948: Finish separating out restconf features"

9 years agoMerge "Bug 2030 : Adding jetty corsfilter in restconf and docgen bundles to avoid...
Ed Warnicke [Wed, 24 Sep 2014 03:39:14 +0000 (03:39 +0000)]
Merge "Bug 2030 : Adding jetty corsfilter in restconf and docgen bundles to avoid cross origin error"

9 years agoBug 1948: Finish separating out restconf features 74/11474/3
Ed Warnicke [Tue, 23 Sep 2014 02:24:47 +0000 (21:24 -0500)]
Bug 1948: Finish separating out restconf features

In order to avoid a maven project cycle in solving
Bug 1948, we needed to separate restconf features.
Note, this is the second step, removing the suffix
 -new.

Change-Id: I3cad2e72141bd78a212dda71b4ef81ba6ab6555a
Signed-off-by: Ed Warnicke <eaw@cisco.com>
9 years agoMerge "Pulled the RPC query out of the data schema loop, so that modules that only...
Ed Warnicke [Wed, 24 Sep 2014 03:04:12 +0000 (03:04 +0000)]
Merge "Pulled the RPC query out of the data schema loop, so that modules that only contain RPCs will also be picked up."

9 years agoBug 2030 : Adding jetty corsfilter in restconf and docgen bundles to avoid cross... 04/11504/3
Harman Singh [Tue, 23 Sep 2014 23:26:58 +0000 (16:26 -0700)]
Bug 2030 : Adding jetty corsfilter in restconf and docgen bundles to avoid cross origin error

Change-Id: I47b860b5c092d54e0ae106b2b832c021502ad52f
Signed-off-by: Harman Singh <harmasin@cisco.com>
9 years agoMerge "Bug 2019: Have FeatureConfigPusher retry installedFeature check until it succe...
Moiz Raja [Tue, 23 Sep 2014 20:06:31 +0000 (20:06 +0000)]
Merge "Bug 2019: Have FeatureConfigPusher retry installedFeature check until it succeeds (or 100ms pass)."

9 years agoBug#2056 odl:xsql karaf command is throwing NPE when empty args 98/11498/1
Sharon Aicler [Tue, 23 Sep 2014 18:40:46 +0000 (11:40 -0700)]
Bug#2056 odl:xsql karaf command is throwing NPE when empty args

Change-Id: I60b6321aea4e02e4a7b409645bc8ebcec9a39112
Signed-off-by: Sharon Aicler <saichler@cisco.com>
9 years agoBug 2019: Have FeatureConfigPusher retry installedFeature check 96/11496/1
Ed Warnicke [Tue, 23 Sep 2014 17:08:42 +0000 (12:08 -0500)]
Bug 2019: Have FeatureConfigPusher retry installedFeature check
until it succeeds (or 100ms pass).

Karaf reports FeatureEvents before it marks the features
installed.  It also only guarantees complete info about
the features when you read the installed ones.  This
is causing a race condition that is at the root of 2019.

So this patch retries at 1ms intervals until we see

Change-Id: I67080475fc4b2dc2f053a286424f676f89164b6e
Signed-off-by: Ed Warnicke <eaw@cisco.com>
9 years agoMerge "Bug 1948: Separate out restconf features."
Devin Avery [Tue, 23 Sep 2014 16:44:11 +0000 (16:44 +0000)]
Merge "Bug 1948: Separate out restconf features."

9 years agoBUG-1521 netconf-impl & netconf-monitoring line coverage. 79/11479/2
Filip Tehlar [Tue, 23 Sep 2014 08:24:06 +0000 (10:24 +0200)]
BUG-1521 netconf-impl & netconf-monitoring line coverage.

Change-Id: Iad7fdaffb3ffeda331f487e2725e2500c6f31c86
Signed-off-by: Filip Tehlar <ftehlar@cisco.com>
9 years agoPulled the RPC query out of the data schema loop, so that 75/11475/1
David K. Bainbridge [Tue, 23 Sep 2014 03:35:45 +0000 (20:35 -0700)]
Pulled the RPC query out of the data schema loop, so that
modules that only contain RPCs will also be picked up.

Signed-off-by: David K. Bainbridge <dbainbri@ciena.com>
Change-Id: I63981d2126209263a762acbc190f4ad15f4dd725

9 years agoMerge "Bug 2026: Fix for Neutron REST API NullPointer exception on POST Port when...
Ed Warnicke [Tue, 23 Sep 2014 02:35:06 +0000 (02:35 +0000)]
Merge "Bug 2026: Fix for Neutron REST API NullPointer exception on POST Port when no fisxed-ips are present"

9 years agoBug 2026: Fix for Neutron REST API NullPointer exception on POST Port when no fisxed... 61/11461/1
Vina Ermagan [Mon, 22 Sep 2014 22:28:00 +0000 (15:28 -0700)]
Bug 2026: Fix for Neutron REST API NullPointer exception on POST Port when no fisxed-ips are present

Change-Id: I80e6435add3e0a07f28dbdf37911cd2bc8b263d0
Signed-off-by: Vina Ermagan <vermagan@cisco.com>
9 years agoBug 963 - OSGi error in Topology manager component - After exiting 50/11450/2
Jan Hajnar [Tue, 16 Sep 2014 13:07:17 +0000 (15:07 +0200)]
Bug 963 - OSGi error in Topology manager component - After exiting
mininet of13 simulation

* added check in onLinkRemoved() to check if link exists before
atempting to delete it (without check, lots of datastore transaction
errors were displayed on mininet shutdown because links were already
removed with nodes)
* added checks to check if node, connector or link exists before trying
to remove it (reads are cheap and delete exception can break the
transaction chain)
* made reads synchronous
* added illegalState exception handling to operation processor
* made affected links removal to be in the same transaction as
Node/NodeConnector removal(creating and applying snapshot for each link
is expensive)
* updated tests

Change-Id: I83b5f91569e64082ce13a4f379cd6261e63fbde6
Signed-off-by: Jan Hajnar <jhajnar@cisco.com>
9 years agoBug 1484 - StatisticManager performance improvement refactoring 05/10605/23
Vaclav Demcak [Wed, 17 Sep 2014 00:53:03 +0000 (02:53 +0200)]
Bug 1484 - StatisticManager performance improvement refactoring

* migration to new MD-SAL API (DataBroker DataChangeListener)
* splitting project api/impl + (javaDoc for api)
* clean pom.xml (xtend dependency and private-package, add sal-binding-broker-impl for test)
* chaining req/resp StatisticNotification per FlowNode per statistic for prevent DS and RPC queue overload
* add multipart msg joiner (StatManager wait for all relevant multipart msg for notification before write relevant
Oper/DS action) - possibility to remove all not used subtrees (target -> reflect actual device contain asap for
all alien node's data).
* add FlowNode registratration - registration needs all SwitchFuture information for make a correct Stat target list (flows,
groups, meters, queues, tables, nrOfTables, ports) - fix for sending not supported notifications to device
* add logic for unique identification FlowId - DeviceFlow (create unique tweens FlowId and flowHash from responded device
flow fields like match, priority, flowCookie, flowCookieMask)
* change FlowCookieMap to FlowHashMap -> List<FlowHash, FlowId>
* implement parallel statistics collecting for FlowCapableNode Sets
* add MockTestSuite for a StatManager workflow interaction testing (neet to be fixed for newest changes)

PASS tests https://jenkins.opendaylight.org/integration/job/integration-master-csit-base-of13
PASS tests https://jenkins.opendaylight.org/integration/job/integration-master-csit-karaf-l2switch-only/
PASS tests https://jenkins.opendaylight.org/integration/job/integration-master-csit-karaf-compatible-min/

Change-Id: I265b57378fc19573d6f18036825cb57f4131ca80
Signed-off-by: Jan Hajnar <jhajnar@cisco.com>
Signed-off-by: Vaclav Demcak <vdemcak@cisco.com>
9 years agoMerge "Bug 1025: Fixed incorrect revision in sal-remote-augment, which caused log...
Ed Warnicke [Mon, 22 Sep 2014 18:31:51 +0000 (18:31 +0000)]
Merge "Bug 1025: Fixed incorrect revision in sal-remote-augment, which caused log polution."

9 years agoMerge "Bug 1309 - Cannot publish LinkDiscovered event"
Tony Tkacik [Mon, 22 Sep 2014 18:29:28 +0000 (18:29 +0000)]
Merge "Bug 1309 - Cannot publish LinkDiscovered event"

9 years agoMerge "Bug 963 - OSGi error in Topology manager component - After exiting mininet...
Tony Tkacik [Mon, 22 Sep 2014 18:28:40 +0000 (18:28 +0000)]
Merge "Bug 963 - OSGi error in Topology manager component - After exiting mininet of13 simulation"

9 years agoBug 1860: increase max permgen in karaf for windows 37/11337/3
Ryan Moats [Thu, 18 Sep 2014 20:52:03 +0000 (15:52 -0500)]
Bug 1860: increase max permgen in karaf for windows

Move file to proper place

Change-Id: Ib9da0ac05b8747ae59aa989c131b9f45d82b2495
Signed-off-by: Ed Warnicke <eaw@cisco.com>
Signed-off-by: Ryan Moats <rmoats@us.ibm.com>
9 years agoBug 1025: Fixed incorrect revision in sal-remote-augment, which caused log polution. 47/11447/1
Tony Tkacik [Mon, 22 Sep 2014 15:12:13 +0000 (17:12 +0200)]
Bug 1025: Fixed incorrect revision in sal-remote-augment, which caused log polution.

Change-Id: I5db5bfb43217d21bad58c913de8dedac5d7cdf23
Signed-off-by: Tony Tkacik <ttkacik@cisco.com>
9 years agoMerge "BUG-1845: implement proper shutdown sequence"
Tony Tkacik [Mon, 22 Sep 2014 14:13:03 +0000 (14:13 +0000)]
Merge "BUG-1845: implement proper shutdown sequence"

9 years agoBUG-2017 Fix possible concurrent modification ex in config subsystem 46/11446/1
Maros Marsalek [Mon, 22 Sep 2014 14:08:45 +0000 (16:08 +0200)]
BUG-2017 Fix possible concurrent modification ex in config subsystem

Change-Id: I879118cee8967edf14e24e8af93945d95bbc9fbe
Signed-off-by: Maros Marsalek <mmarsale@cisco.com>
9 years agoMerge "Bug 1915: Configuration knob for enabling L3 forwarding in OVSDB"
Ed Warnicke [Mon, 22 Sep 2014 14:00:28 +0000 (14:00 +0000)]
Merge "Bug 1915: Configuration knob for enabling L3 forwarding in OVSDB"

9 years agoMerge "Bug 2014: Workaround for karaf listFeatures bug"
Tony Tkacik [Mon, 22 Sep 2014 12:36:53 +0000 (12:36 +0000)]
Merge "Bug 2014: Workaround for karaf listFeatures bug"

9 years agoBug 1915: Configuration knob for enabling L3 forwarding in OVSDB 33/11333/3
Flavio Fernandes [Thu, 18 Sep 2014 19:38:29 +0000 (15:38 -0400)]
Bug 1915: Configuration knob for enabling L3 forwarding in OVSDB

Change-Id: Ic71708b0f4e2dccdfc789d392b8722b68b5ace1f
Signed-off-by: Flavio Fernandes <ffernand@redhat.com>
9 years agoBug 2014: Workaround for karaf listFeatures bug 31/11431/2
Ed Warnicke [Mon, 22 Sep 2014 03:01:42 +0000 (22:01 -0500)]
Bug 2014: Workaround for karaf listFeatures bug

karaf FeatureService.listInstalledFeatures() sometimes
throws a ConcurrentModificationException, this workaround
retries to make sure we succeed.

Change-Id: I4e307d5daba683dfc1b44b6524ce40b3b0e59d1b
Signed-off-by: Ed Warnicke <eaw@cisco.com>
9 years agoBUG-1520 Set correct factory in ImmediateEventExecutorModuleTest 38/11438/1
Maros Marsalek [Mon, 22 Sep 2014 08:02:50 +0000 (10:02 +0200)]
BUG-1520 Set correct factory in ImmediateEventExecutorModuleTest

Change-Id: I664b391e22413514049b55dfb582f3191159a8fc
Signed-off-by: Maros Marsalek <mmarsale@cisco.com>
9 years agoBug 1309 - Cannot publish LinkDiscovered event 37/11437/1
Jan Hajnar [Thu, 11 Sep 2014 13:50:10 +0000 (15:50 +0200)]
Bug 1309 - Cannot publish LinkDiscovered event

* added back MD_SAL_TYPE for nodeConnectorId and nodeId for case when
nodeId and NodeConnectorId are not in openflow format (NodeId and
NodeConnector Ids are strings)
* added test that shows

Change-Id: I73438a14b0b32a9fd3b0bc311663ad29b7bd7dd1
Signed-off-by: Jan Hajnar <jhajnar@cisco.com>
9 years agoMerge "Bug 1850 - The earlier patch missed out on removing the unnecessary call to...
Madhu Venugopal [Mon, 22 Sep 2014 06:27:54 +0000 (06:27 +0000)]
Merge "Bug 1850 - The earlier patch missed out on removing the unnecessary call to INeutronLoadBalancerPoolMemberCRUD. This cleans that up."

9 years agoMerge "Bug 2002 Importing javax.ws.rs.ext explicitly which helps in some cases."
Ed Warnicke [Mon, 22 Sep 2014 03:10:53 +0000 (03:10 +0000)]
Merge "Bug 2002 Importing javax.ws.rs.ext explicitly which helps in some cases."

9 years agoBug 963 - OSGi error in Topology manager component - After exiting 33/11233/6
Jan Hajnar [Tue, 16 Sep 2014 13:07:17 +0000 (15:07 +0200)]
Bug 963 - OSGi error in Topology manager component - After exiting
mininet of13 simulation

* added check in onLinkRemoved() to check if link exists before
atempting to delete it (without check, lots of datastore transaction
errors were displayed on mininet shutdown because links were already
removed with nodes)
* added checks to check if node, connector or link exists before trying
to remove it (reads are cheap and delete exception can break the
transaction chain)
* made reads synchronous
* added illegalState exception handling to operation processor
* made affected links removal to be in the same transaction as
Node/NodeConnector removal(creating and applying snapshot for each link
is expensive)
* updated tests

Change-Id: Icef03a02bbfb88a7cebd0fe35e4b713dcfeb317c
Signed-off-by: Jan Hajnar <jhajnar@cisco.com>
9 years agoMerge "Bug-1978:Leader initializes commit index to last index"
Moiz Raja [Sun, 21 Sep 2014 23:05:46 +0000 (23:05 +0000)]
Merge "Bug-1978:Leader initializes commit index to last index"

9 years agoBug#1961 - Missing recource in the XSQL Karaf Command 41/11341/3
Sharon Aicler [Thu, 18 Sep 2014 21:44:50 +0000 (14:44 -0700)]
Bug#1961 - Missing recource in the XSQL Karaf Command

Change-Id: I718dbd67e6478deea4c42b434bbaa74821bdbbac
Signed-off-by: Sharon Aicler <saichler@cisco.com>
9 years agoBug 1948: Separate out restconf features. 23/11423/1
Ed Warnicke [Thu, 18 Sep 2014 11:59:43 +0000 (06:59 -0500)]
Bug 1948: Separate out restconf features.

In order to avoid a maven project cycle in solving
Bug 1948, we need to separate restconf features.

Note, this is a first step, suffixing everything
with -new.  Subsquently, after everywhere using
odl-restconf has been fixed to use this new repo,
we will deprecate the ones in the mdsal features.xml
and rename these to not have the -new.

Change-Id: I80451e055c071b30a1d0a83ec5a5c14c41afb586
Signed-off-by: Ed Warnicke <eaw@cisco.com>
9 years agoBug-1978:Leader initializes commit index to last index 91/11391/4
Kamal Rameshan [Sat, 20 Sep 2014 00:43:32 +0000 (17:43 -0700)]
Bug-1978:Leader initializes commit index to last index

Change-Id: Ic42a14911e8ff4db40fde2a917c8cb2edf412c76
Signed-off-by: Kamal Rameshan <kramesha@cisco.com>
Signed-off-by: Moiz Raja <moraja@cisco.com>
9 years agoMerge "BUG-1866: sal-compatibility does not register some services"
Ed Warnicke [Sun, 21 Sep 2014 14:32:49 +0000 (14:32 +0000)]
Merge "BUG-1866: sal-compatibility does not register some services"

9 years agoBug 2002 Importing javax.ws.rs.ext explicitly which helps in some cases. 07/11407/2
Madhu Venugopal [Sat, 20 Sep 2014 23:49:54 +0000 (16:49 -0700)]
Bug 2002 Importing javax.ws.rs.ext explicitly which helps in some cases.

We still are trying to figure out the root-cause & this failure has been very random in different setups.

Change-Id: I38da0e23885c2fc4a39d69048c69459f8ac8fc84
Signed-off-by: Madhu Venugopal <mavenugo@gmail.com>
9 years agoBUG-1866: sal-compatibility does not register some services 00/11400/2
Michal Rehak [Sat, 20 Sep 2014 12:11:11 +0000 (14:11 +0200)]
BUG-1866: sal-compatibility does not register some services

- fixed service wiring
- involved statistics collected by statisticsManager (instead fresh request)

Change-Id: I9914dd7e9b20c87d77f7b5350abb9f5c8ba02c0f
Signed-off-by: Michal Rehak <mirehak@cisco.com>
9 years agoBug 1971 : Removing the bounded queue while actor creation to avoid issues in recovery 90/11390/3
Harman Singh [Sat, 20 Sep 2014 00:28:58 +0000 (17:28 -0700)]
Bug 1971 : Removing the bounded queue while actor creation to avoid issues in recovery

Change-Id: I1c423ee0edf2a9b5bc9354f7ddc07150c700d4d3
Signed-off-by: Harman Singh <harmasin@cisco.com>
9 years agoMerge "BUG-1953: fix bad values in netconf monitoring test"
Tony Tkacik [Sat, 20 Sep 2014 14:46:54 +0000 (14:46 +0000)]
Merge "BUG-1953: fix bad values in netconf monitoring test"

9 years agoBUG-1953: fix SAL compatility layer 84/11384/1
Robert Varga [Fri, 19 Sep 2014 19:41:45 +0000 (21:41 +0200)]
BUG-1953: fix SAL compatility layer

Sal compatibility produces pure addresses instead of prefixes as
dictated by the model. Fix moth tests and implementation to emit
prefixes.

Change-Id: I1aca04c0fc22aa973d9578fea33550194c6f9264
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoBUG-1953: fix bad values in netconf monitoring test 83/11383/1
Robert Varga [Fri, 19 Sep 2014 19:40:36 +0000 (21:40 +0200)]
BUG-1953: fix bad values in netconf monitoring test

Uses string values which do not match the format prescribed by the
model. Fix them.

Change-Id: I55efcdd87aea24bcfd805a865319fe324e663768
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoBug 2003: CDS serialization improvements 11/11411/4
tpantelis [Thu, 18 Sep 2014 21:35:53 +0000 (17:35 -0400)]
Bug 2003: CDS serialization improvements

In NormalizedNodeToNodeCodec#encode, significant time was spent
serializing the YangInstanceIdentifier path via PathUtils even though it
wasn't actually needed - the decode method didn't decode it.

This might have been used by WriteModification and MergeModification
originally however they currently serialized/deserialize their
YangInstanceIdentifier path separately from the NormalizedNode via
InstanceIdentifierUtils. It turns out this takes significant time as
well as it's implemented similarly as PathUtils.

So I ended up using NormalizedNodeToNodeCodec to encode/decode the
YangInstanceIdentifier along with the NormalizedNode but changed
InstanceIdentifierUtils to utilize the new PathArgumentSerializer
and the NormalizedNodeSerializer's special QName encoding.

With serializing a 5K batch of WriteModifications with flow data, the
time went down from ~350 ms to ~150 ms. Deserialization was also
improved.

Other smaller optimizations in NormalizedNodeSerializer,
NormalizedNodeType, PathArgumentSerializer and PathArgumentType chopped
another 20-30 ms off the time.

I also changed InstanceIdentifierUtils to serialize/deserialize via the
new PathArgumentSerializer and the NormalizedNodeSerializer's special QName
encoding by default, even when the ID isn't encoded as part of a
NormalizeNode. This seems reasonable to me as a standalone IID will
likely have repeated namespaces and revisions plus we get savings by not
serializing each path arg class name.

Removed the deprecated InstanceIdentifierUtils class in
sal-distributed-datastore bundle.

Change-Id: Iaa29daeaececf4b93065f4d46d0c2796c4d8188f
Signed-off-by: tpantelis <tpanteli@brocade.com>
9 years agoMerge "bug 1888 - FRM Flow Listener registration fail"
Tony Tkacik [Fri, 19 Sep 2014 11:12:32 +0000 (11:12 +0000)]
Merge "bug 1888 - FRM Flow Listener registration fail"

9 years agoMerge "Bug 1963: Fix missing bundle in xsql feature"
Moiz Raja [Fri, 19 Sep 2014 05:35:30 +0000 (05:35 +0000)]
Merge "Bug 1963: Fix missing bundle in xsql feature"

9 years agoBug#1955 - Avoid the bug condition in XSQL 45/11345/2
Sharon Aicler [Thu, 18 Sep 2014 22:58:22 +0000 (15:58 -0700)]
Bug#1955 - Avoid the bug condition in XSQL

Change-Id: I15c8c409ca6bc9aa7c147cdaaa7deca6e640bdcc
Signed-off-by: Sharon Aicler <saichler@cisco.com>
9 years agoBug 1970: Disabling NetconfITSecureTest test 51/11351/1
Ed Warnicke [Fri, 19 Sep 2014 02:38:22 +0000 (21:38 -0500)]
Bug 1970: Disabling NetconfITSecureTest test

Change-Id: Id937c37cd237ee4340a0e8e5bd63561d1e1afe21
Signed-off-by: Ed Warnicke <eaw@cisco.com>
9 years agoBug 1963: Fix missing bundle in xsql feature 28/11328/4
Ed Warnicke [Thu, 18 Sep 2014 15:37:24 +0000 (10:37 -0500)]
Bug 1963: Fix missing bundle in xsql feature

Change-Id: Ieedf3f07bc85110c8c0e4be2d73612311ed5074c
Signed-off-by: Ed Warnicke <eaw@cisco.com>
9 years agoBUG-1845: implement proper shutdown sequence 31/11331/1
Robert Varga [Thu, 18 Sep 2014 17:35:52 +0000 (19:35 +0200)]
BUG-1845: implement proper shutdown sequence

The notification chain should only invoke the successful callback once
all transactions have been committed successfully. This patch does
precisely that by tracking the number of outstanding transactions. Since
this requires notification about successful transactions, too, we get
rid of the unneeded DOMDataCommitErrorInvoker/Listener abstraction.

Change-Id: I38534a3fb79a8a461059504de7b0cdd48348afef
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoMerge "Bug-1903:On recovery all replicated log entries should not be applied to state"
Moiz Raja [Thu, 18 Sep 2014 16:30:45 +0000 (16:30 +0000)]
Merge "Bug-1903:On recovery all replicated log entries should not be applied to state"

9 years agoMerge "Bug 1763: Fixed illegal state in Binding Transaction Chain."
Ed Warnicke [Thu, 18 Sep 2014 16:02:48 +0000 (16:02 +0000)]
Merge "Bug 1763: Fixed illegal state in Binding Transaction Chain."

9 years agoBug 1831 Batch messages on journal recovery 36/11136/11
tpantelis [Sun, 31 Aug 2014 15:15:18 +0000 (11:15 -0400)]
Bug 1831 Batch messages on journal recovery

Added journal log recovery batching support in RaftActor along with
additonal abstract methods for granular recovery control.

It turns out that batching helps a little but the real performance hog
was deserialization. There was a couple inefficient PreConditions in the
NormalizedNodeSerializer (unwanted String concats in the fast path) that
accounted for significant performance degradation. I also made a few
other minor performance enhancements.

Although deserialization is much better with the fixes, I also
implemented some parallelization during shard recovery. I added a
ShardRecoveryCoordinator class that parallelizes deserialization
of journal log enry batches and snapshots for faster recovery time.
The resulting transactions are still committed to the data store in the
order the corresponding snapshot or log batch are received to preserve
data store integrity.

The journal recovery log batch size is configurable vai the config XML.
I also made the shard heartbeat interval and shard snapshot batch count
configurable.
`
Change-Id: I52ef1690bfb6cc486c329ee60f77c52720c24469
Signed-off-by: tpantelis <tpanteli@brocade.com>
9 years agoBUG 1815 - Do not allow Shards to be created till an appropriate schema context is... 46/11346/4
Moiz Raja [Thu, 18 Sep 2014 17:54:47 +0000 (10:54 -0700)]
BUG 1815 - Do not allow Shards to be created till an appropriate schema context is available

The fix works like so,
- ShardManager maintains a list of all the modules that it ever knew about
- ShardManager persists the known modules to disk using persistence
- When ShardManager recovers it reads back the knownModules from persistence
- As ShardManager gets new SchemaContext's it checks whether the modules in
  the new SchemaContext are a superset of the knownModules. If they are then
  ShardManager persists it and let's the Shards know about the new SchemaContext
  otherwise the new SchemaContext is rejected and a message is logged

Also reduced the log level of some log messages in RaftActor from info to debug
it was too verbose

Change-Id: If388f690114c58e6a8df30f34ddac32a99f255e5
Signed-off-by: Moiz Raja <moraja@cisco.com>
9 years agoMerge "Bug 1876 - Added missing parent tag to pom files"
Ed Warnicke [Fri, 19 Sep 2014 19:44:20 +0000 (19:44 +0000)]
Merge "Bug 1876 - Added missing parent tag to pom files"

9 years agoBug 1963 - Fix bad dependency in sal-karaf-xsql 75/11375/1
Devin Avery [Fri, 19 Sep 2014 17:16:14 +0000 (13:16 -0400)]
Bug 1963 - Fix bad dependency in sal-karaf-xsql

Change-Id: I2b9b1d2764f7c65cc2990c715928b05875b12c3c
Signed-off-by: Devin Avery <devin.avery@brocade.com>
9 years agoBUG-1949 Fix race condition in AsyncSshHandler 29/11329/3
Maros Marsalek [Thu, 18 Sep 2014 15:58:48 +0000 (17:58 +0200)]
BUG-1949 Fix race condition in AsyncSshHandler

Remove pending limit due to unexpected behaviour with chunked messages.
Extract Reader/Writer into separate classes.

Also lower the amount of requests sent in SSH Stress integration test.

Change-Id: Idff719ac3a6bed9e8939efa01b8306f2884848fe
Signed-off-by: Maros Marsalek <mmarsale@cisco.com>