| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Jacob Champion <pchampion(at)vmware(dot)com> |
| Cc: | "pryzby(at)telsasoft(dot)com" <pryzby(at)telsasoft(dot)com>, "jcasanov(at)systemguards(dot)com(dot)ec" <jcasanov(at)systemguards(dot)com(dot)ec>, "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: pg_upgrade parallelism |
| Date: | 2021-11-23 21:43:25 |
| Message-ID: | 61285.1637703805@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Jacob Champion <pchampion(at)vmware(dot)com> writes:
> Right. What I'm worried about is, if disk space or write performance on
> the new cluster is a concern, then having a copy-mode upgrade silently
> use copy-on-write could be a problem if the DBA needs copy mode to
> actually copy.
Particularly for the cross-filesystem case, where it would not be
unreasonable to expect that one could dismount or destroy the old FS
immediately afterward. I don't know if recent kernels try to make
that safe/transparent.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Robert Haas | 2021-11-23 21:44:43 | Re: Post-CVE Wishlist |
| Previous Message | Tom Lane | 2021-11-23 21:41:06 | Re: Sequence's value can be rollback after a crashed recovery. |