https://jira.opendaylight.org This file is an XML representation of an issue en-us 8.20.10 820010 22-06-2022 https://jira.opendaylight.org/browse/CONTROLLER-1671 controller <p>ShardDataTree's transaction pipelining can trigger a long chain of immediate completions, which can eat eat up the stack as demonstrated in <a href="https://jira.opendaylight.org/browse/CONTROLLER-1620" title="StackOverflowError while shard is committing" class="issue-link" data-issue-key="CONTROLLER-1620"><del>CONTROLLER-1620</del></a>.</p> <p>In that particular case we are effectively dealing with list traversal (pending transactions in startPreCommit) being done one entry at a time, with each entry eating into the stack:</p> <p> at org.opendaylight.controller.cluster.datastore.ShardDataTree.processNextPending(ShardDataTree.java:762)<br/> at org.opendaylight.controller.cluster.datastore.ShardDataTree.processNextPendingTransaction(ShardDataTree.java:707)<br/> at org.opendaylight.controller.cluster.datastore.ShardDataTree.startPreCommit(ShardDataTree.java:828)<br/> at org.opendaylight.controller.cluster.datastore.SimpleShardDataTreeCohort.preCommit(SimpleShardDataTreeCohort.java:115)<br/> at org.opendaylight.controller.cluster.datastore.ChainedCommitCohort.preCommit(ChainedCommitCohort.java:63)<br/> at org.opendaylight.controller.cluster.datastore.FrontendReadWriteTransaction.successfulDirectCanCommit(FrontendReadWriteTransaction.java:212)<br/> at org.opendaylight.controller.cluster.datastore.FrontendReadWriteTransaction$5.onSuccess(FrontendReadWriteTransaction.java:200)<br/> at org.opendaylight.controller.cluster.datastore.FrontendReadWriteTransaction$5.onSuccess(FrontendReadWriteTransaction.java:197)<br/> at org.opendaylight.controller.cluster.datastore.SimpleShardDataTreeCohort.successfulCanCommit(SimpleShardDataTreeCohort.java:183)<br/> at org.opendaylight.controller.cluster.datastore.ShardDataTree.lambda$processNextPendingTransaction$0(ShardDataTree.java:718)<br/> at org.opendaylight.controller.cluster.datastore.ShardDataTree.processNextPending(ShardDataTree.java:762)<br/> at org.opendaylight.controller.cluster.datastore.ShardDataTree.processNextPendingTransaction(ShardDataTree.java:707)<br/> at org.opendaylight.controller.cluster.datastore.ShardDataTree.startPreCommit(ShardDataTree.java:828)<br/> at org.opendaylight.controller.cluster.datastore.SimpleShardDataTreeCohort.preCommit(SimpleShardDataTreeCohort.java:115)<br/> at org.opendaylight.controller.cluster.datastore.ChainedCommitCohort.preCommit(ChainedCommitCohort.java:63)<br/> at org.opendaylight.controller.cluster.datastore.FrontendReadWriteTransaction.successfulDirectCanCommit(FrontendReadWriteTransaction.java:212)<br/> at org.opendaylight.controller.cluster.datastore.FrontendReadWriteTransaction$5.onSuccess(FrontendReadWriteTransaction.java:200)<br/> at org.opendaylight.controller.cluster.datastore.FrontendReadWriteTransaction$5.onSuccess(FrontendReadWriteTransaction.java:197)<br/> at org.opendaylight.controller.cluster.datastore.SimpleShardDataTreeCohort.successfulCanCommit(SimpleShardDataTreeCohort.java:183)<br/> at org.opendaylight.controller.cluster.datastore.ShardDataTree.lambda$processNextPendingTransaction$0(ShardDataTree.java:718)<br/> at org.opendaylight.controller.cluster.datastore.ShardDataTree.processNextPending(ShardDataTree.java:762)</p> <p>I believe this can be refactored so that we have a simple while() loop executing canCommit/preCommit(), which communicate back their immediate result. The loop would bail as soon as it encounters an async operation.</p> <p>Operating System: All<br/> Platform: All</p> CONTROLLER-1671

Rework transaction cohorts to better support immediate completion

Improvement Confirmed Unresolved Unassigned Robert Varga Fri, 12 May 2017 14:23:27 +0000 Tue, 25 Jul 2023 08:24:36 +0000 clustering 0 0 Development External issue ID 8433 External issue URL Issue Type Rank 0|i02s0f: