[ 
https://issues.apache.org/jira/browse/HBASE-30486?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

ASF GitHub Bot updated HBASE-30486:
-----------------------------------
    Labels: pull-request-available  (was: )

> Fix XOR used instead of exponentiation in VerifyReplication backoff
> -------------------------------------------------------------------
>
>                 Key: HBASE-30486
>                 URL: https://issues.apache.org/jira/browse/HBASE-30486
>             Project: HBase
>          Issue Type: Bug
>          Components: mapreduce
>            Reporter: Xiao Liu
>            Assignee: Xiao Liu
>            Priority: Major
>              Labels: pull-request-available
>             Fix For: 4.0.0-alpha-1, 2.7.0, 3.1.0, 3.0.1, 2.6.8, 2.5.17
>
>
> {\{VerifyReplicationRecompareRunnable}} calculates the retry delay using:
> {code:java}
> sleepMs = sleepMs * (2 ^ reCompareBackoffExponent);
> {code}
> Java's \{{^}} operator performs XOR, not exponentiation. With exponent 
> \{{2}}, the multiplier becomes \{{0}}, removing the delay between subsequent 
> retries and potentially reporting bad rows before replication catches up.
> Replace XOR with exponentiation to restore the intended backoff behavior.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to