[
https://issues.apache.org/jira/browse/PHOENIX-7838?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Viraj Jasani updated PHOENIX-7838:
----------------------------------
Fix Version/s: 5.3.2
(was: 5.3.1)
> LogicalTableNameBaseIT wrap admin.snapshot() in retry helper
> ------------------------------------------------------------
>
> Key: PHOENIX-7838
> URL: https://issues.apache.org/jira/browse/PHOENIX-7838
> Project: Phoenix
> Issue Type: Sub-task
> Components: test
> Reporter: Andrew Kyle Purtell
> Assignee: Andrew Kyle Purtell
> Priority: Minor
> Fix For: 5.4.0, 5.3.2
>
>
> Four {{LogicalTableNameBaseIT}} helpers call {{admin.snapshot()}} directly
> with a snapshot name deterministically derived from the table name, so RPC
> retries or concurrent forks produce duplicate in-flight requests that HMaster
> rejects with {{SnapshotCreationException}} "already running another snapshot
> on the same table", even though the original procedure still succeeds. The
> test bypasses {{ConnectionQueryServicesImpl.createSnapshot}} so the fix adds
> a {{snapshotWithRetry(admin, name, tableName)}} helper and routes all four
> call sites through it.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
