| From: | Michael Paquier <michael(dot)paquier(at)gmail(dot)com> |
|---|---|
| To: | Pavan Deolasee <pavan(dot)deolasee(at)gmail(dot)com> |
| Cc: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Heikki Linnakangas <hlinnaka(at)iki(dot)fi>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: FSM corruption leading to errors |
| Date: | 2016-10-20 05:20:54 |
| Message-ID: | CAB7nPqR13+6a0_ZPznjMqNrWpzF+BBfQCWXJiYpCDv3yxG704g@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Thu, Oct 20, 2016 at 2:11 PM, Pavan Deolasee
<pavan(dot)deolasee(at)gmail(dot)com> wrote:
> 4. Remove FSM and VM files for the affected tables (I don't think if it's
> safe to do this on a running server)
Definitely not while the server is running... For VMs a good way would
be to use pg_visibility's pg_truncate_visibility_map(), but only for
9.6~. For FSM there is no real solution, and actually a
pg_truncate_fsm would prove to be useful here. So users would need to
stop once the server if there are corrupted VMs or FSMs, delete them
manually, and then run VACUUM to rebuild them.
--
Michael
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Pavan Deolasee | 2016-10-20 05:50:40 | Re: FSM corruption leading to errors |
| Previous Message | Michael Paquier | 2016-10-20 05:14:13 | Re: Password identifiers, protocol aging and SCRAM protocol |