| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Bruce Momjian <bruce(at)momjian(dot)us> |
| Cc: | pgsql-committers(at)postgresql(dot)org |
| Subject: | Re: pgsql: Fix building of large (bigger than shared_buffers) hash indexes. |
| Date: | 2016-06-27 21:35:38 |
| Message-ID: | 7612.1467063338@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-committers |
Bruce Momjian <bruce(at)momjian(dot)us> writes:
> On Fri, Jun 24, 2016 at 08:57:47PM +0000, Tom Lane wrote:
>> Fix building of large (bigger than shared_buffers) hash indexes.
> Do we have any way of helping people find out if they need to recreate
> their hash indexes?
I do not think that's much of an issue. The failure mode is hardly
non-obvious, because index searches would NEVER find any pre-existing
rows. (Well, there'd be circa one chance in 2^32 of an accidental
hashcode match, but that certainly won't be enough to mask the fact
that the index is broken.) Anyway, users of hash indexes are probably
accustomed to needing to reindex, on account of no WAL support :-(
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Robert Haas | 2016-06-27 21:57:26 | pgsql: Fix mistakes in pg_visibility documentation. |
| Previous Message | Peter Geoghegan | 2016-06-27 21:30:53 | Re: pgsql: Fix building of large (bigger than shared_buffers) hash indexes. |