[
https://issues.apache.org/jira/browse/HBASE-22190?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16831955#comment-16831955
]
Zach York commented on HBASE-22190:
-----------------------------------
[~Apache9] Why did you merge PR 211 for this while I already had a solution up
in PR 209? I had already addressed your initial comments.
Some concerns with the approach taken in PR 211 for 2 reasons:
1. It completely removes the performance improvement of the modification time.
This means that for each call of refreshCache it will loop through all
Snapshots and read all the manifests. Maybe it isn't that much of benefit?
2. It moves the synchronized to the run method which to me seems wrong. It is
not safeguarding what we truly want to lock. If tomorrow we had a new method
that called refreshCache we could run into this again. I understand you added
this so that you can clear the cache in the case of an IOE.
> SnapshotFileCache may fail to load the correct snapshot file list when there
> is an on-going snapshot operation
> --------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-22190
> URL: https://issues.apache.org/jira/browse/HBASE-22190
> Project: HBase
> Issue Type: Bug
> Components: snapshots
> Reporter: Duo Zhang
> Priority: Blocker
> Fix For: 3.0.0, 2.2.0, 2.3.0, 2.1.5
>
>
> And it seems that it is not only a test issue, we do delete the files under
> the archive directory, which is incorrect.
> Need to find out why, this maybe a serious bug.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
