[
https://issues.apache.org/jira/browse/ASTERIXDB-3556?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17923219#comment-17923219
]
ASF subversion and git services commented on ASTERIXDB-3556:
------------------------------------------------------------
Commit 7a46963be8b1ffaf66f97c2b0794af5f9bddd98c in asterixdb's branch
refs/heads/master from Murtadha Makki Al Hubail
[ https://gitbox.apache.org/repos/asf?p=asterixdb.git;h=7a46963be8 ]
[ASTERIXDB-3556][STO] Ensure local recovery is run on resume
- user model changes: no
- storage format changes: no
- interface changes: yes
Details:
- Set the system state to CORRUPTED on resume to ensure
local recovery is run.
- Ensure cached files clean up is performed all the time
when we have any cached files.
Ext-ref: MB-65062
Change-Id: I6600823447f478fc70a22cf27a9967235570562c
Reviewed-on: https://asterix-gerrit.ics.uci.edu/c/asterixdb/+/19373
Tested-by: Jenkins <[email protected]>
Reviewed-by: Murtadha Hubail <[email protected]>
Reviewed-by: Ritik Raj <[email protected]>
> Cached files clean up is missed on healthy system state
> -------------------------------------------------------
>
> Key: ASTERIXDB-3556
> URL: https://issues.apache.org/jira/browse/ASTERIXDB-3556
> Project: Apache AsterixDB
> Issue Type: Bug
> Components: STO - Storage
> Affects Versions: 0.9.9
> Reporter: Murtadha Makki Al Hubail
> Assignee: Murtadha Makki Al Hubail
> Priority: Major
> Labels: triaged
> Fix For: 0.9.10
>
>
> When the node reports its system state as healthy, the cached files clean up
> is skipped. This can result in the cache and the remote storage files
> diverging in cases where the remote storage files changed since the node last
> cached those files. Similarly, local recovery is also skipped on healthy
> system state and this can result in files not cleaned up after they are
> downloaded from the remote storage.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
