| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | bricklen <bricklen(at)gmail(dot)com> |
| Cc: | "pgsql-general(at)postgresql(dot)org" <pgsql-general(at)postgresql(dot)org> |
| Subject: | Re: ERROR: invalid page in block 1226710 of relation base/16750/27244 |
| Date: | 2015-10-22 16:15:56 |
| Message-ID: | 48933.1445530556@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
bricklen <bricklen(at)gmail(dot)com> writes:
> There are several hot standby servers attached to the master, some
> streaming, and one in a different data centre that is using WAL shipping
> only.
> The streaming slave IIRC got the corruption from the master (I can't check
> now, it was rebuilt).
> What would have happened to the WAL-shipping-only standby if the WALs were
> all applied? Would it have it balked at applying a WAL containing bad data
> from the master, or would it have applied the WAL and continued on? For the
> latter, would physical corruption on the master even transfer via WAL?
Hard to tell. I'd have guessed that corruption that made a page
unreadable would not transfer across WAL (streaming or otherwise), because
the master could not have read it in to apply an update to it. However,
we don't know the exact sequence of events here; there may have more than
one step on the way to disaster.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | bricklen | 2015-10-22 16:25:36 | Re: ERROR: invalid page in block 1226710 of relation base/16750/27244 |
| Previous Message | Tom Lane | 2015-10-22 16:12:33 | Re: carray_to_bytea? |