| From: | Thomas Munro <thomas(dot)munro(at)enterprisedb(dot)com> |
|---|---|
| To: | Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> |
| Cc: | Andrew Gierth <andrew(at)tao11(dot)riddles(dot)org(dot)uk>, Pg Hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Checkpoint not retrying failed fsync? |
| Date: | 2018-04-08 09:17:52 |
| Message-ID: | CAEepm=23GCvc+sY_Y-HH=d06iyYiKFG_W7KXn76aRpuFiAP2aQ@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Sun, Apr 8, 2018 at 5:36 PM, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
> Won't in the success case, you need to delete each member (by
> something like bms_del_member) rather than just using bms_free?
Thanks for looking at this. Yeah, if requests for segment numbers 0
and 1 were in "requests", and 0 succeeded but then 1 fails, my
previous patch would leave both in there to be retried next time
around. I thought that was pretty harmless so I didn't worry about it
before, but of course you're right that it's not necessary to retry
the ones that succeeded, so we could remove them as we go. New patch
attached.
--
Thomas Munro
http://www.enterprisedb.com
| Attachment | Content-Type | Size |
|---|---|---|
| 0001-Make-sure-we-don-t-forget-about-fsync-requests-af-v3.patch | application/octet-stream | 2.6 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Andreas Karlsson | 2018-04-08 09:41:06 | Re: PostgreSQL's handling of fsync() errors is unsafe and risks data loss at least on XFS |
| Previous Message | Pavel Stehule | 2018-04-08 07:38:43 | Re: csv format for psql |