mirrors/onos
1
0
Fork 0
mirror of https://github.com/opennetworkinglab/onos.git synced 2025-10-23 21:31:00 +02:00
Code Issues Projects Releases Wiki Activity
7,182 Commits 69 Branches 240 Tags
Commit Graph

1436 Commits

Author SHA1 Message Date
Yuta HIGUCHI
c9438f14dd SQueue: lazily deserialize iterator 
Change-Id: I09ba574a0479181774e287d8a39286c976740675
 2014-11-30 18:13:37 +00:00
alshabib
978d2417da add iterator support to SQueue 
Change-Id: I83df62326e54368e40fa15df522c05ba2dc67ff6
 2014-11-29 15:29:46 -08:00
Yuta HIGUCHI
47b2f55b29 lower log levels 
Change-Id: I48feba520fe72b531d3b6c36309216c549d599ca
 2014-11-28 20:13:15 -08:00
Yuta HIGUCHI
82e5326b23 For ONOS-356 
- Add current InstallationFuture information on TimeoutException
- Set timeout values propotional to batch size
- Fix for ConcurrentModificationException
- Check if src/dst is part of the graph before path computation

Change-Id: Iabeac7939c52502b83bf9ebcbe2023539de3ae99
 2014-11-28 19:00:56 -08:00
alshabib
9c57bddf00 check if there are any stats remaining when removing 
Change-Id: I6af100ff73d5f577f38901b8f0ec881e51253cc8

apply fix to distributed store

Change-Id: I829394e919aa79f10e01ac42904f5708b296cf51
 2014-11-28 19:22:09 -05:00
Brian O'Connor
72a034c863 Initial implementation of distributed intent batch queue 
Change-Id: I7ffed03651569ade1be1e8dca905bfaf369b7e03
 2014-11-28 22:42:44 +00:00
Thomas Vachuska
a68be8199f Fixed a null pointer exception in SimpleMastershipStore; need to verify whether same exists in distributed one. 
Change-Id: I89ad124114586eb69bb9de3702085e8587cb315c
 2014-11-28 21:18:06 +00:00
Yuta HIGUCHI
f0f2dfc1df tablet-leader command to check current Raft leader 
Change-Id: Id360db21988a50c3e2895c5194d59b0ba4cb49e4
 2014-11-26 20:54:41 -08:00
Yuta HIGUCHI
f1f2ac051d Construct GraphDescription with online/active Device/Link only 
Change-Id: I9312c0c8ae190bf0200bd040823b6f7a3e7a15e1
 2014-11-27 04:08:49 +00:00
Yuta HIGUCHI
b6eb9144f2 DistributedFlowRuleStore: handle no master present case 
Change-Id: I56f4bce549aecc689d7a4798485b025a87ecdeb9
 2014-11-27 00:42:39 +00:00
Thomas Vachuska
6b7920ddaf Measuring topology performance. Fixing a defect that caused us to run w/o accumulator when config file was not present. 
Change-Id: I5ad538b8a441cd6ff2aefea49a0def10b8e0f4d5
 2014-11-26 09:02:50 +00:00
Madan Jampani
4bb70c60ce Fix for copycat log max size setting. Turns out ^ is the bitwise exlusive or :) 
Change-Id: I5a8e5e367d9fddc7441814aed91008608583aab4
 2014-11-25 23:47:36 -08:00
Yuta HIGUCHI
80d5659844 GossipStores: remove potentially blocking method out of netty thread 
Change-Id: I2da9ba745c3a63bf9709fb77c1f260ea8f4529a8
 2014-11-25 20:44:10 -08:00
Yuta HIGUCHI
ae1d210a8f Add Collections.emptySet() 
Change-Id: I51e845ed69f688a859089a221643fd6d04ac7afa
 2014-11-25 20:44:09 -08:00
Yuta HIGUCHI
0136562edc DatabaseManager: move retry timeout out from finally clause 
Change-Id: If4760bdaa8ac64d5db80c7b6700d75a89099f179
 2014-11-25 20:44:09 -08:00
Thomas Vachuska
e4b6bb2515 Intents are now removed after being withdrawn. 
Change-Id: I7574fe94add00abf58c71c6122bb3dc5aafa0f79
 2014-11-26 04:31:14 +00:00
Yuta HIGUCHI
f6171233a9 register DefaultGroupId to Kryo 
Change-Id: I650482164acf4b97f1bcd1a1633806a3b526a14d
 2014-11-26 00:01:33 +00:00
Yuta HIGUCHI
0658627cc1 Run Anti-Entropy in background 
Change-Id: I233185d15f52359899427e214339be44cb62971c
 2014-11-25 14:27:27 -08:00
Yuta HIGUCHI
b6cfac39b4 GossipStores: iterate using forEach 
Change-Id: I1c230932ab1c9112692708a4d5c869eac3c50db8
 2014-11-25 22:07:13 +00:00
Ray Milkey
e97ede9548 ONOS-239 Retrigger compilation when new resources are available 
When an intent is withdrawn and frees up resources, trigger
recompilation of any failed intents to allow the ones waiting
for resources to be installed.

Change-Id: Ic15678378ce41516a7eab890b4b4898aeb901f78
 2014-11-25 21:52:58 +00:00
Yuta HIGUCHI
2f1583342c DistributedFlowRuleStore: register FlowRuleEvent 
Change-Id: I1f904a283c8e619a6ec9094d62ddfba44a41c119
 2014-11-25 13:32:06 -08:00
Yuta HIGUCHI
4423acc4aa GossipHostStore: prep for making initial size configurable 
Change-Id: I256f46ff9772015dac2f02678c5396ee2b566741
 2014-11-25 12:34:44 -08:00
Yuta HIGUCHI
d88400b288 DatabaseManager: extended election Timeout 2s -> 5s 
Change-Id: I192113f634cd27c21b7d987b784bc81b1c6bde44
 2014-11-25 12:14:39 -08:00
Yuta HIGUCHI
e367fb9718 Add probes to DistributedIntentStore 
Change-Id: I23a5823d3924392dc17166404a17fc1918c01453
 2014-11-25 06:56:15 +00:00
Brian O'Connor
520c052e36 Changing Intent Ids to use explicit id assignment 
Change-Id: I5a4bff87842c37a869e7691b353529eaefc929db
 2014-11-25 05:49:35 +00:00
Yuta HIGUCHI
dfa45c1563 DistributedFlowRuleStore: support remote removeFlowRule needed for cancelling Batch 
Change-Id: I40f8dd8c2008e93c5ac7393295374726f83353c7
 2014-11-25 03:22:11 +00:00
Jonathan Hart
7466d61447 Null checking in packet-out mechanism. 
If the switch has recently disconnected, we can get some NullPointerExceptions.

Change-Id: Ia5a854ed5cd6a933974c855e7297faa542626d9e
 2014-11-24 17:09:53 -08:00
Madan Jampani
d1a467fd6f Make leadership election more robust to failures. 
- Catch exceptions thrown by lock extension calls.
    - Dealing with potential race conditions between joining and withdrawing from a race.

Change-Id: I429045b33f5972c459d5ed031fe8593438813e8d
 2014-11-24 16:58:11 -08:00
Yuta HIGUCHI
45e5cd1778 Keep retrying until Database table is ready. 
Change-Id: Idd696e3807435c65a422064e078cf5993a216df1
 2014-11-24 15:42:21 -08:00
Yuta HIGUCHI
c8f302698f Enable Raft version of IntentStore 
- Turning off state transition verification for now

Change-Id: I59d9796995d7349cd19e741c6f00e7dd2d2c3d73
 2014-11-24 10:13:07 -08:00
Brian O'Connor
895c12acaa Fix for HZ intent store 
The old value should be 'null'

Change-Id: I55d7826faa2c24bc911e6cc85be7d0e8bdc69c36
 2014-11-23 19:21:48 -08:00
Michele Santuari
4a33807957 SinglePoint to MultiPoint Intent initial implementation 
Change-Id: I1010997ce4ea993ae34afb8dab4b6c0ae112448d
 2014-11-23 18:09:11 -08:00
Yuta HIGUCHI
4e45081d22 DatabaseManager: larger raft log size limit 
Change-Id: Ib93b11ce55d9c038fad8e4fabfa8efc089899656
 2014-11-23 16:17:21 -08:00
Yuta HIGUCHI
bcac4995df Flatten MastershipTermService 
Change-Id: Ic75b7376f7436c76f69127d6137577ec2fc09d2b
 2014-11-23 16:17:07 -08:00
Brian O'Connor
427a1760d4 Updating Intent Manager to deal with failures. 
Added ids to Flow batch futures.
Adding some basic unit tests for IntentManger
Adding failedIds to the completedOperation in FlowRuleManager

Change-Id: I7645cead193299f70d319d254cd1e82d96909e7b
 2014-11-23 16:16:34 -08:00
Pavlin Radoslavov
af529d8f92 Adding-back "Wait for a raft quorum + logging a message to inform about the wait" 
This reverts commit 17319764e20309e9be9c00ffe15ad76d22f8a757.

Change-Id: I6e5e4d5ef5e06930c80ca96a4e1e9de7d71eee70
 2014-11-22 21:03:42 -08:00
Yuta HIGUCHI
03e77e975a ClusterMessagingProtocol: stop processing in netty handler thread 
- Fix for io.netty.util.concurrent.BlockingOperationException

Change-Id: Ie0f4dee2c3a49aa4b03674f6f7678f32fcf07a44
 2014-11-22 19:35:42 -08:00
Yuta HIGUCHI
91768e30e5 Kryo related fixes 
- KryoNamespace to allow control over registration id

Change-Id: Idc2a0e27a09916657c725ee97e4366109144cc66
 2014-11-22 19:35:28 -08:00
Yuta HIGUCHI
51b8f6acba DistributedIntentStore: remove transient state, after transition to parking state 
Change-Id: Id0a5bf6a94c03c7eb15d399fd85bf00f4322f6d4
 2014-11-22 11:34:44 -08:00
Madan Jampani
6234fd44e7 Updates to ClusterMessagigProtocolClient's handling of remote node connectivity issues. 
Change-Id: If3cd220bef339cc57b2a5d034c6e86bad2202a9f
 2014-11-22 00:54:30 -08:00
Yuta HIGUCHI
89a7f47115 Distributed IntentStore family: relax state transition check 
- allow INSTALLED -> INSTALLED
- allow WITHDRWAN -> SUBMITTED

Change-Id: I7ba9c7c6e29b39ce005fb15bfd68feb1751cb0af
 2014-11-21 20:59:28 -08:00
Pavlin Radoslavov
17319764e2 Revert "Wait for a raft quorum + logging a message to inform about the wait" 
Reverting-back for further investigation.

This reverts commit d2f457b80f9bea8c34f895d822e2328281386a6b.

Change-Id: I83d2c171fd26e4008c0fd36ed635993eb18b3718
 2014-11-21 19:03:42 -08:00
Madan Jampani
d2f457b80f Wait for a raft quorum + logging a message to inform about the wait 
Change-Id: I6b1e5a6f2a0ca57e0a435877068ba83bc9311953
 2014-11-21 18:39:23 -08:00
Madan Jampani
2b6ca91b59 Fix for recovering node availability after a temporary downtime. Currently we can only recover once. 
Change-Id: I556273fa49aa745f0b742454a0529b8fa1251575
 2014-11-21 14:44:45 -08:00
Madan Jampani
ddaffd0fd8 1. Lock.lockAsync does not throw DatabaseException. 
2. Changed thread pool in LockManager to be non-static.

Change-Id: Ie4e9acd497bacb9d6d812836a930ee79f92cf555
 2014-11-21 14:24:30 -08:00
Madan Jampani
1ee91780ea LeadershipService: Support for a leaderBoard. 
Change-Id: I0dd8267e104466ec65a2c67d23d1c4d923cad266

Change-Id: I6bc548510400eacabb12482f8fba1b7f2abb0604
 2014-11-20 21:27:23 -08:00
Yuta HIGUCHI
d2499435c8 ClusterMessagingProtocolClient: transition to not connected state on IO error 
Change-Id: Iac0af5b5a55868d2677aecf18e63e00018d5113f
 2014-11-20 18:46:36 -08:00
Yuta HIGUCHI
2805222cc7 DatabaseManager: workaround for larger ONOS cluster 
- sleep before retrying if listTable failed due to non-Timeout error
- do not throw exception even if retry max reached

Change-Id: I5764894f0830c11b07d63aefbd0bbb10fe41af74
 2014-11-20 18:35:45 -08:00
Yuta HIGUCHI
1a012723c6 Lower log level 
Change-Id: I41c72ef6bba5bf59f76ab08c09fb44cf9a5bbae5
 2014-11-20 17:47:09 -08:00
Madan Jampani
1d3494ec8f Initial cut at Leadership Manager 
Change-Id: I658c6fca3dc6f686e0f7facc9e7b443679ebae1e

Change-Id: I293906add41ff4310e3584847d806345e0312703

Change-Id: I7fb13a72ba4aef10d7c2262b96e0df64efecfcef
 2014-11-20 14:37:49 -08:00