| From: | Heikki Linnakangas <hlinnaka(at)iki(dot)fi> |
|---|---|
| To: | Thomas Munro <thomas(dot)munro(at)gmail(dot)com> |
| Cc: | pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | pendingOps table is not cleared with fsync=off |
| Date: | 2020-05-08 21:21:18 |
| Message-ID: | 3cbc7f4b-a5fa-56e9-9591-c886deb07513@iki.fi |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hi!
I noticed that commit 3eb77eba5a changed the logic in
ProcessSyncRequests() (formerly mdsync()) so that if you have fsync=off,
the entries are not removed from the pendingOps hash table. I don't
think that was intended.
I propose the attached patch to move the test for enableFsync so that
the entries are removed from pendingOps again. It looks larger than it
really is because it re-indents the block of code that is now inside the
"if (enableFsync)" condition.
- Heikki
| Attachment | Content-Type | Size |
|---|---|---|
| clear-pendingOps-with-fsync-off-1.patch | text/x-patch | 6.1 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Andres Freund | 2020-05-08 21:27:45 | Re: refactoring basebackup.c |
| Previous Message | Alvaro Herrera | 2020-05-08 21:09:16 | Re: 2pc leaks fds |