controller.git
9 years agoBug 1029: Remove dead code: sal-binding-spi 08/7208/1
Ed Warnicke [Sun, 18 May 2014 23:01:43 +0000 (18:01 -0500)]
Bug 1029: Remove dead code: sal-binding-spi

opendaylight/md-sal/sal-binding-spi is dead code that is not being
built and hasn't for some time.  Its complicating keeping
versions up to date and verifying that they are up to date.
If it needs to be revived in the future it can be retrieved
from git.

Change-Id: I2698f4bfa687e24e65989c1257c5f80d84ebe262
Signed-off-by: Ed Warnicke <eaw@cisco.com>
9 years agoChange in sal.connectionservice.notifyNodeDisconnectFromMaster 39/7139/1
Alessandro Boch [Sat, 17 May 2014 20:40:14 +0000 (13:40 -0700)]
Change in sal.connectionservice.notifyNodeDisconnectFromMaster

- On node disconnect from Master, notify the respective protocol plugin, not all of them

Change-Id: I2bfb30d2f649a1bc3f07be60150a814bcd1b8a45
Signed-off-by: Alessandro Boch <aboch@cisco.com>
9 years agoMerge "Bug 967: Do not translate leaf Instance Identifiers."
Ed Warnicke [Fri, 16 May 2014 15:15:24 +0000 (15:15 +0000)]
Merge "Bug 967: Do not translate leaf Instance Identifiers."

9 years agoBug 967: Do not translate leaf Instance Identifiers. 11/7111/1
Tony Tkacik [Fri, 16 May 2014 11:02:33 +0000 (13:02 +0200)]
Bug 967: Do not translate leaf Instance Identifiers.

Improved extraction of Instance Identifiers from
DOM Data Change Event.

Logs in console are poluted with DeserializationException,
which fails to deserialize DOM Instance Identifier
to Binding Instance Identifier.

Added explicit check against schema context to
see if Instance Identifier is representable
in Binding Format, if not we will not report
is as part of subtree Data Change event.

Change-Id: Iaee2390fd089e0a2d66ec658aab4e7dab2a93a46
Signed-off-by: Tony Tkacik <ttkacik@cisco.com>
9 years agoBug 1022 - Add ability to lookup dependent Module's attribute. 10/7110/1
Tomas Olvecky [Fri, 16 May 2014 11:24:49 +0000 (13:24 +0200)]
Bug 1022 - Add ability to lookup dependent Module's attribute.

Add 2 methods to DependencyResolver to allow modules look up values
of dependent bean attributes.

Change-Id: I2ebee82cc1260a84a16f5e69b32c9234bf576e87
Signed-off-by: Tomas Olvecky <tolvecky@cisco.com>
9 years agoMerge "Add lispflowmapping specific configuration options"
Giovanni Meo [Fri, 16 May 2014 08:24:09 +0000 (08:24 +0000)]
Merge "Add lispflowmapping specific configuration options"

9 years agoMerge "Node Name in non-default container should default to Node Name in Default...
Andrew Kim [Thu, 15 May 2014 22:00:43 +0000 (22:00 +0000)]
Merge "Node Name in non-default container should default to Node Name in Default Slice"

9 years agoNode Name in non-default container should default to Node Name in Default Slice 04/7004/2
Pramila Singh [Wed, 14 May 2014 22:35:13 +0000 (15:35 -0700)]
Node Name in non-default container should default to Node Name in Default Slice

Change-Id: I81670c824bfcb4d7f400ef224abf329b6bc20abd
Signed-off-by: Pramila Singh <pramisin@cisco.com>
9 years agoMerge "Bug 509: Improve conflict detection of leaf merges"
Tony Tkacik [Thu, 15 May 2014 15:50:38 +0000 (15:50 +0000)]
Merge "Bug 509: Improve conflict detection of leaf merges"

9 years agoMerge changes Ibe0145be,I0764bfa9
Tony Tkacik [Thu, 15 May 2014 15:50:08 +0000 (15:50 +0000)]
Merge changes Ibe0145be,I0764bfa9

* changes:
  BUG-509: remove DataAndMetadataSnapshot.Builder.getDataTree()
  BUG-509: remove StoreMetadataNode.children()

9 years agoMerge "BUG-509: remove StoreUtils.initialChangeEvent()"
Tony Tkacik [Thu, 15 May 2014 15:49:44 +0000 (15:49 +0000)]
Merge "BUG-509: remove StoreUtils.initialChangeEvent()"

9 years agoMerge "BUG-509: StoreUtils should operate on NormalizedNode"
Tony Tkacik [Thu, 15 May 2014 15:49:11 +0000 (15:49 +0000)]
Merge "BUG-509: StoreUtils should operate on NormalizedNode"

9 years agoBug 509: Improve conflict detection of leaf merges 22/7022/1
Robert Varga [Thu, 15 May 2014 14:55:02 +0000 (16:55 +0200)]
Bug 509: Improve conflict detection of leaf merges

Original conflict detection for merge action on leaves
only looked at node version and subtree version, which
caused the transaction to fail even if the leaf was
unmodified and contained the same value.

Conflict detection now checks if the leaves are not same,
and only in that case it checks for concurrent writes.

Change-Id: I5e43233e86adf96e3dad2a93cdcf512b3d17977a
Signed-off-by: Tony Tkacik <ttkacik@cisco.com>
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoBUG-509: remove DataAndMetadataSnapshot.Builder.getDataTree() 17/7017/1
Robert Varga [Thu, 15 May 2014 12:37:19 +0000 (14:37 +0200)]
BUG-509: remove DataAndMetadataSnapshot.Builder.getDataTree()

This getter and its associated field is completely unused. Remove it to
improve clarity.

Change-Id: Ibe0145bee610c7a2b147852f43d0492ca5061adc
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoMerge "Bug 1002 - REST POST transaction is not commited"
Tony Tkacik [Thu, 15 May 2014 12:02:38 +0000 (12:02 +0000)]
Merge "Bug 1002 - REST POST transaction is not commited"

9 years agoMerge "Cosmetics: fix typo facilitate"
Ed Warnicke [Thu, 15 May 2014 11:12:31 +0000 (11:12 +0000)]
Merge "Cosmetics: fix typo facilitate"

9 years agoBug 1002 - REST POST transaction is not commited 90/6990/4
Vaclav Demcak [Wed, 14 May 2014 12:39:10 +0000 (14:39 +0200)]
Bug 1002 - REST POST transaction is not commited

fix REST POST create workflow (create data only), fix according to specification
  http://tools.ietf.org/html/draft-bierman-netconf-restconf-03#section-2.4

I'm missing implementation for ietf-restconf:errors, so I can't do more for now
  http://tools.ietf.org/html/draft-bierman-netconf-restconf-03#page-48

* fix a check for an existing node in DataStore
* add warn log and a correct ResponseException
* fix tests for BrokerFacade

Change-Id: I68c95e831f42cae4df0608a40c585251a631693f
Signed-off-by: Vaclav Demcak <vdemcak@cisco.com>
9 years agoBUG-509: remove StoreMetadataNode.children() 15/7015/1
Robert Varga [Thu, 15 May 2014 09:10:36 +0000 (11:10 +0200)]
BUG-509: remove StoreMetadataNode.children()

This patch removes the now-unneeded access, which means the children
iteration order is no longer part of StoreMetadataNode, allowing us to
use more memory-efficient HashMap instead of LinkedHashMap. Let's do
precisely that.

Change-Id: I0764bfa9861f9df4d3014151db96edf1024e3f9b
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoBUG-509: remove StoreUtils.initialChangeEvent() 14/7014/1
Robert Varga [Thu, 15 May 2014 08:56:48 +0000 (10:56 +0200)]
BUG-509: remove StoreUtils.initialChangeEvent()

This method is unused. Remove it, such that StoreUtils are now
completely independent of org.opendaylight.controller.md.sal, making
them a candidate for move to yang-data-*.

Change-Id: Idf945aee6fd4a80f8ca58fc50bcc997880a60ec0
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoBUG-509: StoreUtils should operate on NormalizedNode 13/7013/1
Robert Varga [Thu, 15 May 2014 08:47:00 +0000 (10:47 +0200)]
BUG-509: StoreUtils should operate on NormalizedNode

This prunes the use of MetadataNode from StoreUtils. It turns out it is
completely unnecessary. This removes the last user of
Metadata.getChildren(), which will be removed in a subsequent patch.

Change-Id: I778e77b826576b8fa53e040a825d615dab7d253f
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoBug 967: Fixed incorrect delete was addressing augmentation of augmentation 82/6782/2
Tony Tkacik [Wed, 7 May 2014 11:10:40 +0000 (13:10 +0200)]
Bug 967: Fixed incorrect delete was addressing  augmentation of augmentation

DataNormalizationOperation was constructed incorrectly, if augmentation
was already augmented, which leaded to incorrect construction of instance
identifier for delete.

Patchset 2: Fixed normalization in cases of mixin inside mixin
e.g. augmentation containing choice with cases.

Change-Id: I580e5dfe41ce504775ac454b66368037f4d707bb
Signed-off-by: Tony Tkacik <ttkacik@cisco.com>
9 years agoMerge "BUG 1001 Failing test - RestGetOperationTest.getStreamsTest"
Tony Tkacik [Wed, 14 May 2014 10:16:45 +0000 (10:16 +0000)]
Merge "BUG 1001 Failing test - RestGetOperationTest.getStreamsTest"

9 years agoMerge "Add Pagination to Neutron Networks Northbound API"
Giovanni Meo [Wed, 14 May 2014 09:45:10 +0000 (09:45 +0000)]
Merge "Add Pagination to Neutron Networks Northbound API"

9 years agoMerge "BUG-372 Add reconnecting to netconf-connector for unexpected session drop"
Tony Tkacik [Wed, 14 May 2014 09:02:00 +0000 (09:02 +0000)]
Merge "BUG-372 Add reconnecting to netconf-connector for unexpected session drop"

9 years agoBUG 1001 Failing test - RestGetOperationTest.getStreamsTest 84/6984/1
Jozef Gloncak [Wed, 14 May 2014 08:23:24 +0000 (10:23 +0200)]
BUG 1001 Failing test - RestGetOperationTest.getStreamsTest

responseBody variable now contained some data and therefore it wasn't xml
tag <streams/> but <streams></streams>

Change-Id: I73004f848f0dfc5f2929f8f8ea2dbd6afc25d35a
Signed-off-by: Jozef Gloncak <jgloncak@cisco.com>
9 years agoMerge "REST API to fetch Node Property"
Andrew Kim [Tue, 13 May 2014 18:30:11 +0000 (18:30 +0000)]
Merge "REST API to fetch Node Property"

9 years agoMerge "Bug 617: Remove extend files from sal-rest-connector"
Tony Tkacik [Tue, 13 May 2014 15:02:41 +0000 (15:02 +0000)]
Merge "Bug 617: Remove extend files from sal-rest-connector"

9 years agoBUG-372 Add reconnecting to netconf-connector for unexpected session drop 34/6934/1
Maros Marsalek [Tue, 13 May 2014 14:10:59 +0000 (16:10 +0200)]
BUG-372 Add reconnecting to netconf-connector for unexpected session drop

Netconf connector will reattempt connection if the session to remote device is dropped.
Increase sleep factor in netconf connector from 1.0 to 1.5.
This prevents connection attempt every second in case remote device is not comming back for longer time.
Connection attempts span increases gradually.

Change-Id: Icc4b993942e9cc6897c0da858eb867afa4c8aa87
Signed-off-by: Maros Marsalek <mmarsale@cisco.com>
9 years agoMerge "Bug 809: Enhancements to the toaster example"
Tony Tkacik [Tue, 13 May 2014 13:52:24 +0000 (13:52 +0000)]
Merge "Bug 809: Enhancements to the toaster example"

9 years agoBUG-981: Fixed incorrect condition in ResolveDataChangeEventsTask 29/6929/1
Robert Varga [Tue, 13 May 2014 11:52:10 +0000 (13:52 +0200)]
BUG-981: Fixed incorrect condition in ResolveDataChangeEventsTask

When we insert a new subtree, we need to make sure to call serialization
on the entire, such that any new elements which appear have their
classed attached to paths.

This reuses a previous code path, which pruned the walk tree as soon as
it encountered a node which did not have any listeners attached (thus
improving performance when we're gathering listeners).

In this specific case, though, this check is not correct: we need to
walk the path because it may be placing new data in the subtree.

At the same time add debug statements which made pinpointing this
problem possible.

Change-Id: I1687ba45817a84ea4a888496a90defc5896216c9
Signed-off-by: Tony Tkacik <ttkacik@cisco.com>
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoREST API to fetch Node Property 15/6915/1
Sapan Shah [Tue, 13 May 2014 00:06:28 +0000 (17:06 -0700)]
REST API to fetch Node Property

Change-Id: Iad693deefcc3a3f5420e8e19ddb81919c788738f
Signed-off-by: Sapan Shah <sapshah@cisco.com>
9 years agoMerge "Bug-574 Added a 4th RPC Invocation strategy for the case where the RPC has...
Tony Tkacik [Mon, 12 May 2014 18:48:31 +0000 (18:48 +0000)]
Merge "Bug-574 Added a 4th RPC Invocation strategy for the case where the RPC has an output, but no input"

9 years agoMerge "BUG-704 Remove pax from netconf identity-ref test."
Tony Tkacik [Mon, 12 May 2014 16:35:07 +0000 (16:35 +0000)]
Merge "BUG-704 Remove pax from netconf identity-ref test."

9 years agoBug-574 Added a 4th RPC Invocation strategy for the case where the RPC has an output... 90/6190/2
Mayank Agarwal [Mon, 14 Apr 2014 21:58:31 +0000 (14:58 -0700)]
Bug-574 Added a 4th RPC Invocation strategy for the case where the RPC has an output, but no input

Change-Id: I7d1a596f3bb1362044cc240a15bdb9bb7d094c55
Signed-off-by: Mayank Agarwal <mayagarw@cisco.com>
9 years agoMerge "Bug 508: Improved error reporting for failed canCommit phase."
Ed Warnicke [Mon, 12 May 2014 14:41:49 +0000 (14:41 +0000)]
Merge "Bug 508: Improved error reporting for failed canCommit phase."

9 years agoMerge "Bug 981: Fixed accidental shaddowing of data in Data Change Event."
Giovanni Meo [Mon, 12 May 2014 13:53:28 +0000 (13:53 +0000)]
Merge "Bug 981: Fixed accidental shaddowing of data in Data Change Event."

9 years agoBug 508: Improved error reporting for failed canCommit phase. 91/6891/3
Tony Tkacik [Mon, 12 May 2014 10:33:58 +0000 (12:33 +0200)]
Bug 508: Improved error reporting for failed canCommit phase.

Original implementation did not provide enough information
when transaction precondition failed.

Introduced new Exception DataPreconditionFailedException
which captures fail reason and path to subtree
for which precondition failed.

This helps debug accidental overwrites / conflicting
writes between transactions.

Change-Id: I0133fd3fdb8e6492bf457dab7b20efa67fd6e2af
Signed-off-by: Tony Tkacik <ttkacik@cisco.com>
9 years agoMerge "Bug 849: Fixed NPE in Translated Data Change Events."
Giovanni Meo [Mon, 12 May 2014 11:05:43 +0000 (11:05 +0000)]
Merge "Bug 849: Fixed NPE in Translated Data Change Events."

9 years agoBUG-704 Remove pax from netconf identity-ref test. 34/6834/2
Maros Marsalek [Fri, 9 May 2014 12:58:08 +0000 (14:58 +0200)]
BUG-704 Remove pax from netconf identity-ref test.

Make this test a regular unit test without pax.

Change-Id: I0f9767d860f25bb60494ded2011e3f398fbcf43f
Signed-off-by: Maros Marsalek <mmarsale@cisco.com>
9 years agoBug 981: Fixed accidental shaddowing of data in Data Change Event. 83/6883/4
Tony Tkacik [Sun, 11 May 2014 20:46:24 +0000 (22:46 +0200)]
Bug 981: Fixed accidental shaddowing of data in Data Change Event.

Not all Normalized DOM InstanceIdentifier has their counterpart
in Binding-Aware representation. Added explicit check to
detect mixin instance-identifiers which could lead to
incorrect deserialization into maps and shadowing parent data.

Change-Id: I66166f50c636ce0dfcd962efac7fc488da622336
Signed-off-by: Tony Tkacik <ttkacik@cisco.com>
9 years agoMerge "BUG-648: Improve performance by reusing old data nodes"
Tony Tkacik [Mon, 12 May 2014 07:05:37 +0000 (07:05 +0000)]
Merge "BUG-648: Improve performance by reusing old data nodes"

9 years agoMerge "BUG-980: stop emiting copyright"
Tony Tkacik [Mon, 12 May 2014 07:04:12 +0000 (07:04 +0000)]
Merge "BUG-980: stop emiting copyright"

9 years agoMerge "BUG-731: RoutedRegistration.close() should not throw"
Tony Tkacik [Mon, 12 May 2014 07:03:49 +0000 (07:03 +0000)]
Merge "BUG-731: RoutedRegistration.close() should not throw"

9 years agoMerge "Remove overriding versions"
Tony Tkacik [Mon, 12 May 2014 07:03:30 +0000 (07:03 +0000)]
Merge "Remove overriding versions"

9 years agoMerge "Unify maven-bundle-plugin version at 2.4.0"
Tony Tkacik [Mon, 12 May 2014 07:03:15 +0000 (07:03 +0000)]
Merge "Unify maven-bundle-plugin version at 2.4.0"

9 years agoMerge "BUG-972: correct Precondition"
Tony Tkacik [Mon, 12 May 2014 07:02:55 +0000 (07:02 +0000)]
Merge "BUG-972: correct Precondition"

9 years agoMerge "Add serialVersionUID to Serializable"
Tony Tkacik [Mon, 12 May 2014 07:02:33 +0000 (07:02 +0000)]
Merge "Add serialVersionUID to Serializable"

9 years agoMerge "BUG-868: stop using InstanceIdentifier.builder()"
Tony Tkacik [Mon, 12 May 2014 07:02:17 +0000 (07:02 +0000)]
Merge "BUG-868: stop using InstanceIdentifier.builder()"

9 years agoBUG-648: Improve performance by reusing old data nodes 86/6886/2
Robert Varga [Thu, 8 May 2014 16:29:30 +0000 (18:29 +0200)]
BUG-648: Improve performance by reusing old data nodes

This patchset gives yang-data-api the chance to optimize copy
operations we execute when creating a new generation.

Change-Id: Ib23a87e116a0a04f7fd4f2b709d7239e213ea744
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoAdd serialVersionUID to Serializable 70/6870/2
Robert Varga [Sun, 11 May 2014 07:16:29 +0000 (09:16 +0200)]
Add serialVersionUID to Serializable

Serializable classes have to have serialVersionUID. Add it where it's
missing.

Change-Id: I23689005878280b6ee9f97fb12ae0b15151f20d7
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoBUG-868: stop using InstanceIdentifier.builder() 69/6869/3
Robert Varga [Sun, 11 May 2014 06:41:18 +0000 (08:41 +0200)]
BUG-868: stop using InstanceIdentifier.builder()

The static helper method has been deprecated. Use either the non-static
one, or child()/augmentation() helpers.

Change-Id: Ic69057d2db669153dde60ba1797e8139623a755c
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoBUG-981: remove deprecated use of constructors 68/6868/2
Robert Varga [Sun, 11 May 2014 05:35:34 +0000 (07:35 +0200)]
BUG-981: remove deprecated use of constructors

This removes the use of all constructors, such that we initialize in one
go.

Change-Id: Iff461a878fb7061b0ea1c8efea6a34b833645784
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoMerge "BUG-893: do not store classPool"
Tony Tkacik [Sun, 11 May 2014 16:46:53 +0000 (16:46 +0000)]
Merge "BUG-893: do not store classPool"

9 years agoMerge "BUG-642: get rid of sneakyThrow"
Tony Tkacik [Sun, 11 May 2014 16:46:25 +0000 (16:46 +0000)]
Merge "BUG-642: get rid of sneakyThrow"

9 years agoCosmetics: fix typo facilitate 66/6866/1
Flavio Fernandes [Sun, 11 May 2014 03:27:02 +0000 (23:27 -0400)]
Cosmetics: fix typo facilitate

Change-Id: I351e285dd128f1d633f4940624096348b065f16a
Signed-off-by: Flavio Fernandes <ffernand@redhat.com>
9 years agoBug 935 Fixed, added order key in instruction and action as flow yang model expects it. 39/6639/2
Amit Mandke [Thu, 1 May 2014 20:48:55 +0000 (13:48 -0700)]
Bug 935 Fixed, added order key in instruction and action as flow yang model expects it.

Change-Id: Ib7a82db71f4915496574ec2b8731cb0161351309
Signed-off-by: Amit Mandke <ammandke@cisco.com>
9 years agoBUG-893: do not store classPool 55/6855/1
Robert Varga [Sat, 10 May 2014 13:43:45 +0000 (15:43 +0200)]
BUG-893: do not store classPool

This prevents storing a reference to the classPool. It is not necessary
and it makes thinking about accesses much easier.

Change-Id: Iebbdac46c3ca445bbe36ba6efd19d3150d728f7f
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoBUG-642: get rid of sneakyThrow 53/6853/1
Robert Varga [Sat, 10 May 2014 11:25:21 +0000 (13:25 +0200)]
BUG-642: get rid of sneakyThrow

This patch rids us of unneeded usage of Exceptions.sneakyThrow. It
generally hides problems and is only useful for generated (or sloppy)
code.

Change-Id: Ibb73f4546b1785c4bcb3e6c033249f58efd1d2d6
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoBUG-980: stop emiting copyright 27/6827/3
Robert Varga [Fri, 9 May 2014 07:33:44 +0000 (09:33 +0200)]
BUG-980: stop emiting copyright

This is a stop-gap quickfix to stop doing the wrong thing.

Change-Id: I51a2ca012ea605632a737788005cf4143d9c1f44
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoBUG-891: do not drop events silently 36/6836/1
Robert Varga [Fri, 9 May 2014 15:22:19 +0000 (17:22 +0200)]
BUG-891: do not drop events silently

As it turns out, infrastructure silently drops things on the floor if
exceptions happen. When these exceptions occur, they hint at major
problems. Promote tham to warnings.

Change-Id: Ib4afdc7ff1a17e562211b3cfbb30735a592d0a2d
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoBUG-731: RoutedRegistration.close() should not throw 30/6830/1
Robert Varga [Fri, 9 May 2014 10:16:50 +0000 (12:16 +0200)]
BUG-731: RoutedRegistration.close() should not throw

Routed RPC implementations should be allowed to unregister themselves
without incurring a checked exceptions. This allows downstream users to
elide having to deal with raw Exception.

Change-Id: Ia356c6272da3d22a375c954e192f79786ab42395
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoMerge "BUG-969 Move fake read on nodes/node to separate thread."
Tony Tkacik [Fri, 9 May 2014 08:06:58 +0000 (08:06 +0000)]
Merge "BUG-969 Move fake read on nodes/node to separate thread."

9 years agoMerge "Making it easier for cygwin users to run OpenDaylight using run.sh. With this...
Giovanni Meo [Fri, 9 May 2014 07:45:34 +0000 (07:45 +0000)]
Merge "Making it easier for cygwin users to run OpenDaylight using run.sh. With this change, the users will only have to add the following line to their .bashrc to make it run: export CLASSPATH=".;lib/*;""

9 years agoBUG-969 Move fake read on nodes/node to separate thread. 26/6826/1
Maros Marsalek [Fri, 9 May 2014 07:27:55 +0000 (09:27 +0200)]
BUG-969 Move fake read on nodes/node to separate thread.

This will prevent blocking of config thread.

Change-Id: Ieb2e1c192ef91775153837956e6e6a428985c2d1
Signed-off-by: Maros Marsalek <mmarsale@cisco.com>
9 years agoBug 809: Enhancements to the toaster example 79/6379/7
tpantelis [Fri, 18 Apr 2014 19:10:47 +0000 (15:10 -0400)]
Bug 809: Enhancements to the toaster example

See https://bugs.opendaylight.org/show_bug.cgi?id=809 for details.

Need someone to commit this!

Change-Id: Icb9697de77d844d816ec61a0e7a6ac2100d185ae
Signed-off-by: tpantelis <tpanteli@brocade.com>
9 years agoRemove overriding versions 18/6818/1
Robert Varga [Thu, 8 May 2014 17:12:17 +0000 (19:12 +0200)]
Remove overriding versions

We have declaration overriding versions from parent, triggering eclipse
warnings. Remove them.

Change-Id: Id5ef1cf34e9c8dfb4550a5b0240fb6b950d05f7f
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoBUG-972: correct Precondition 17/6817/1
Robert Varga [Thu, 8 May 2014 15:30:25 +0000 (17:30 +0200)]
BUG-972: correct Precondition

This is just a tiny adjustment of a Precondition, which should have been
triggered before this access. isPresent() returns true/false, which will
be autoboxed, so will always pass checkNotNull. use checkArgument()
instead, which will ensure that isPresent() == true.

Change-Id: I53fe6913c7a0f8073a193bc76c0792059c26fe2a
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoUnify maven-bundle-plugin version at 2.4.0 16/6816/1
Robert Varga [Thu, 8 May 2014 11:47:16 +0000 (13:47 +0200)]
Unify maven-bundle-plugin version at 2.4.0

This removes superfluous version declarations, fixing about 70 eclipse
warnings. At the same time it bumps the version 2.4.0, to unify it -- we
were using both 2.3.7 and 2.4.0 at the same time.

Change-Id: I0f3f6b68c5c05cb9cc226f75b4eb4dab7d58651d
Signed-off-by: Robert Varga <rovarga@cisco.com>
9 years agoMerge "Bug-590: Packet loss on first time ping test"
Giovanni Meo [Wed, 7 May 2014 23:21:44 +0000 (23:21 +0000)]
Merge "Bug-590: Packet loss on first time ping test"

9 years agoMerge "Upgrade enunciate version to 1.28 -Convert XmlElementRef to XmlElement to...
Alessandro Boch [Wed, 7 May 2014 23:20:15 +0000 (23:20 +0000)]
Merge "Upgrade enunciate version to 1.28   -Convert XmlElementRef to XmlElement to allow doc generation to work properly   -Add swagger module dependency to enunciate"

9 years agoUpgrade enunciate version to 1.28 33/6733/4
mrajvaid [Mon, 5 May 2014 23:10:56 +0000 (16:10 -0700)]
Upgrade enunciate version to 1.28
  -Convert XmlElementRef to XmlElement to allow doc generation to work properly
  -Add swagger module dependency to enunciate

Change-Id: Ia41e7733bd44965951ca83aaa7738e2df74151da
Signed-off-by: mrajvaid <mrajvaid@cisco.com>
9 years agoFelix Web Console authenticates using HTTP-Bridge, which is similar to web-gui 58/6558/3
mrajvaid [Mon, 28 Apr 2014 19:34:25 +0000 (12:34 -0700)]
Felix Web Console authenticates using HTTP-Bridge, which is similar to web-gui

By default web console perform its own authentication using default user and passowrd
If user password is modified via web, felix console can not authenticate

Change-Id: I38a277ffabdfac652325b83fe80facffc043c669
Signed-off-by: mrajvaid <mrajvaid@cisco.com>
9 years agoBug-590: Packet loss on first time ping test 98/6798/2
Flavio Fernandes [Wed, 7 May 2014 18:00:51 +0000 (14:00 -0400)]
Bug-590: Packet loss on first time ping test

Bump severity of log in exception code path.
Undo copy and paste mistake, which caused unwanted line left behind.

Change-Id: If73e516635172e31e794cd1524611c1cdb15f89f
Signed-off-by: Flavio Fernandes <ffernand@redhat.com>
9 years agoBug:969 - Add fake read to block thread to sync with BI broker. 97/6797/1
Tomas Olvecky [Wed, 7 May 2014 17:54:42 +0000 (19:54 +0200)]
Bug:969 - Add fake read to block thread to sync with BI broker.

Block netconf client until Binding Independent broker loads inventory
models.

Change-Id: I8b2bcb82bcdf97598665b0b45f8fdfed19072e3c
Signed-off-by: Tomas Olvecky <tolvecky@cisco.com>
9 years agoMerge "Fix netconf-connector filter construction"
Tony Tkacik [Wed, 7 May 2014 16:46:15 +0000 (16:46 +0000)]
Merge "Fix netconf-connector filter construction"

9 years agoMerge "Add missing exception to log message in SchemaAwareRpcBroker"
Tony Tkacik [Wed, 7 May 2014 16:46:08 +0000 (16:46 +0000)]
Merge "Add missing exception to log message in SchemaAwareRpcBroker"

9 years agoFix netconf-connector filter construction 96/6796/2
Maros Marsalek [Wed, 7 May 2014 15:32:30 +0000 (17:32 +0200)]
Fix netconf-connector filter construction

Change-Id: I9b36458d0e3bb2a14b303e14ddcd8dcab4e7b33d
Signed-off-by: Maros Marsalek <mmarsale@cisco.com>
9 years agoAdd Pagination to Neutron Networks Northbound API 05/6005/4
Dave [Wed, 9 Apr 2014 15:28:19 +0000 (16:28 +0100)]
Add Pagination to Neutron Networks Northbound API

This commit adds pagination to the neutron network northbound API.
It is based on the information at the following link.
http://docs.openstack.org/api/openstack-network/2.0/content/pagination.html

In addition to not providing a "next" link on the last page of results this
implementation does not have a "previous" link on the first page.

Change-Id: I2c048c9cdc907d759e62fb8850b76cfa293b7fd2
Signed-off-by: Dave Tucker <dave.j.tucker@hp.com>
9 years agoMerge "Initial mdsal and config karaf features"
Tony Tkacik [Wed, 7 May 2014 15:43:12 +0000 (15:43 +0000)]
Merge "Initial mdsal and config karaf features"

9 years agoMerge "Increase mdsal's notification queue size, make it configurable."
Tony Tkacik [Wed, 7 May 2014 15:42:46 +0000 (15:42 +0000)]
Merge "Increase mdsal's notification queue size, make it configurable."

9 years agoInitial mdsal and config karaf features 53/6753/3
Ed Warnicke [Tue, 6 May 2014 14:57:45 +0000 (09:57 -0500)]
Initial mdsal and config karaf features

This loads cleanly into karaf :)

Instructions for checking:

cd controller/

mvn clean install

cd opendaylight/distributions/opendaylight-karaf/target/assembly/bin
sh karaf

at the karaf prompt

feature:install odl-mdsal-restconf
feature:list -i

(odl-mdsal-restconf will pull everything else through).

PatchSet2:  Fixed failure to use odl- prefix in features for -all

Change-Id: Ib85347eaab7729d1b2723eb7ce59a3c18f2eef0e
Signed-off-by: Ed Warnicke <eaw@cisco.com>
9 years agoAdd missing exception to log message in SchemaAwareRpcBroker 94/6794/1
Maros Marsalek [Wed, 7 May 2014 15:13:00 +0000 (17:13 +0200)]
Add missing exception to log message in SchemaAwareRpcBroker

Change-Id: I1e8196babd9f5486fbe1edadba5f610f00d1ac37
Signed-off-by: Maros Marsalek <mmarsale@cisco.com>
9 years agoMerge "FRM should use async function on southbound event"
Giovanni Meo [Wed, 7 May 2014 15:08:07 +0000 (15:08 +0000)]
Merge "FRM should use async function on southbound event"

9 years agoMerge "Bug-590: Packet loss on first time ping test"
Giovanni Meo [Wed, 7 May 2014 15:06:38 +0000 (15:06 +0000)]
Merge "Bug-590: Packet loss on first time ping test"

9 years agoMerge "Bug 933: Fixed incorrect update type when Link was deleted."
Ed Warnicke [Wed, 7 May 2014 14:38:53 +0000 (14:38 +0000)]
Merge "Bug 933: Fixed incorrect update type when Link was deleted."

9 years agoFRM should use async function on southbound event 54/6754/2
Alessandro Boch [Tue, 6 May 2014 17:39:18 +0000 (10:39 -0700)]
FRM should use async function on southbound event

- On reacting to southbound events (switch up/down, ports added/removed, ...)
  FRM is supposed to use asynchronous calls for flows installation/removal.
  This is needed because these events may affect a large number of flows,
  for which we do not want to slow down the processing via sync calls.
  Also, logic has been we never do process synchronously a southbound event.
- In case of port up, FRM was still using a synchronous call.
  This patch is for taking care of that.

Change-Id: I4b6dae0df75ad778ea2f1d555a258a48ea66c57b
Signed-off-by: Alessandro Boch <aboch@cisco.com>
9 years agoBug 967: Statistics Manager should delete flow, not augmentation only. 83/6783/3
Tony Tkacik [Wed, 7 May 2014 11:15:26 +0000 (13:15 +0200)]
Bug 967: Statistics Manager should delete flow, not augmentation only.

Change-Id: I0c74e3e55f61078851e28b27847945c0ea8a9c51
Signed-off-by: Tony Tkacik <ttkacik@cisco.com>
9 years agoIncrease mdsal's notification queue size, make it configurable. 87/6787/1
Tomas Olvecky [Wed, 7 May 2014 12:56:56 +0000 (14:56 +0200)]
Increase mdsal's notification queue size, make it configurable.

Add System property mdsal.notificationqueue.size to allow tweaking
of queue size.
Increase default queue size from 10 to 1000.

Change-Id: I91e8ac3939092af2b26a7a4d952fe561557b7a46
Signed-off-by: Tomas Olvecky <tolvecky@cisco.com>
9 years agoBug 933: Fixed incorrect update type when Link was deleted. 85/6785/1
Tony Tkacik [Wed, 7 May 2014 11:51:45 +0000 (13:51 +0200)]
Bug 933: Fixed incorrect update type when Link was deleted.

Change-Id: I660fc6006785bcf7560a0479ca29f38ceaa208b6
Signed-off-by: Tony Tkacik <ttkacik@cisco.com>
9 years agoMerge "Bug 946: Fixed eclipse errors with nagasena projects"
Tony Tkacik [Wed, 7 May 2014 11:14:22 +0000 (11:14 +0000)]
Merge "Bug 946: Fixed eclipse errors with nagasena projects"

9 years agoMerge "Fixed typo in SnapshotBackedWriteTransaction class"
Tony Tkacik [Wed, 7 May 2014 11:14:00 +0000 (11:14 +0000)]
Merge "Fixed typo in SnapshotBackedWriteTransaction class"

9 years agoBug 617: Remove extend files from sal-rest-connector 21/6821/6
tpantelis [Sun, 4 May 2014 17:29:22 +0000 (13:29 -0400)]
Bug 617: Remove extend files from sal-rest-connector

Converted ControllerContext and BrokerFacade xtend code to java.

Addressed review comments:
  - Removed SupressWarnings "all"
  - Converted use of xext classes to guava
  - Removed sneakyThrow

Patch set 4: mistake - no changes

Patch set 5: Converted RestconfImpl

Patch set 6: removed xtend plugin anf dependencies from pom file

Change-Id: I45c22b22fee07a178faba9fcb9e52d3ff12a6697
Signed-off-by: tpantelis <tpanteli@brocade.com>
9 years agoMerge "Breaking up 6360: dummy console"
Giovanni Meo [Tue, 6 May 2014 19:12:42 +0000 (19:12 +0000)]
Merge "Breaking up 6360: dummy console"

9 years agoMerge "Fixed opendaylight-karaf to have resources in target/assembly"
Giovanni Meo [Tue, 6 May 2014 17:12:05 +0000 (17:12 +0000)]
Merge "Fixed opendaylight-karaf to have resources in target/assembly"

9 years agoFixed opendaylight-karaf to have resources in target/assembly 50/6750/1
Ed Warnicke [Tue, 6 May 2014 16:03:03 +0000 (11:03 -0500)]
Fixed opendaylight-karaf to have resources in target/assembly

Change-Id: I1a5ad76bb2c6e6a49cad347df14a1f6630e64475
Signed-off-by: Ed Warnicke <eaw@cisco.com>
9 years agoBug-590: Packet loss on first time ping test 74/6674/2
Flavio Fernandes [Sun, 4 May 2014 01:44:34 +0000 (21:44 -0400)]
Bug-590: Packet loss on first time ping test

In this commit, we are addressing a race that could cause the exception (ie first ip packet) to
be lost. What is observed with this bug is that the non-arp packet -- in this case the icmp packet --
is being 'exceptioned' to the SimpleForwarding code path because the node has not yet been programmed
with a flow that would cause it to forward the packet to its destination w/out involving the controller.

The race in this case is caused by Simpleforwarding handling the packet before HostTracker is fully aware
of the destination host. In such case, the forwarding fails to determine where to send the packet and drops it.

The proposed change in this commit takes a stab at handling the race by keeping the packet until either
1) the destination is known by HostTracker or 2) a max time expires. In order to accomplish that, the change
includes the creation of a thread that will periodically age out any packet that have been pending for too
long. The change also adds a concurrent map for efficient lookup among the pending packets. There is a
limitation in this implementation: it will keep at most 1 packet pending per destination ip address. This is
believed not to be a problem, because of the short amount of time it takes for hostTracker to learn about the
destination after the packet is put int the pending map.

Change-Id: I7ec1de69f7841eb7ca5603715b54685be9781401
Signed-off-by: Flavio Fernandes <ffernand@redhat.com>
9 years agoBUG-509 transaction future result listening 43/6743/1
Michal Rehak [Tue, 6 May 2014 13:09:27 +0000 (15:09 +0200)]
BUG-509 transaction future result listening

- replaced transactionResult.get() with future listener

Change-Id: I6f047a3c65fbb2cf669cb8912920aacc0c2848d3
Signed-off-by: Michal Rehak <mirehak@cisco.com>
9 years agoAdd lispflowmapping specific configuration options 41/6741/1
Lorand Jakab [Tue, 6 May 2014 12:12:30 +0000 (15:12 +0300)]
Add lispflowmapping specific configuration options

Change-Id: I02bd8515423c5b4e8ddaf1c1b7f70c9f4b84e2a6
Signed-off-by: Lorand Jakab <lojakab@cisco.com>
9 years agoBug 849: Fixed NPE in Translated Data Change Events. 40/6740/1
Tony Tkacik [Tue, 6 May 2014 11:31:05 +0000 (13:31 +0200)]
Bug 849: Fixed NPE in Translated Data Change Events.

Change-Id: I7f79b9a713414192f14d934fbc3b9eabb3a2a051
Signed-off-by: Tony Tkacik <ttkacik@cisco.com>
9 years agoMerge "Bug: 627 Added RESTConf API Explorer that dynamically generates API documentat...
Tony Tkacik [Tue, 6 May 2014 06:52:40 +0000 (06:52 +0000)]
Merge "Bug: 627 Added RESTConf API Explorer that dynamically generates API documentation and serves it through swagger UI. Details are documented in this wiki: https://wiki.opendaylight.org/view/OpenDaylight_Controller:MD-SAL:Restconf_API_Explorer"