Re: Something is rotten in the state of Denmark...

On Thu, Apr 2, 2015 at 2:40 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Robert Haas <robertmhaas(at)gmail(dot)com> writes:
>> On Thu, Apr 2, 2015 at 12:54 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>>> It looks to me like an appropriate fix would be as attached; thoughts?
>
>> Hmm, that fix doesn't reach as far as what I did. My proposal would
>> regard a catalog snapshot as immediately stale, so if we're asked for
>> a catalog snapshot multiple times before InitPostgres() is called,
>> we'll take a new one every time. Your proposal invalidates them just
>> once, after setting up MyDatabaseId. Assuming yours is safe, it's
>> better, because it invalidates less aggressively.
>
> Right.
>
>> The only thing I'm worried about is that I think
>> PerformAuthentication() runs before InitPostgres(); sinval isn't
>> working at all at that point, even for shared catalogs.
>
> No, PerformAuthentication is called by InitPostgres.

Oops, OK.

> However, I'm having second thoughts about whether we've fully diagnosed
> this. Three out of the four failures we've seen in the buildfarm reported
> "cache lookup failed for access method 403", not "could not open relation
> with OID 2601" ... and I'm so far only able to replicate the latter
> symptom. It's really unclear how the former one could arise, because
> nothing that vacuum.sql does would change xmin of the rows in pg_am.

It probably changes the *relfilenode* of pg_am, because it runs VACUUM
FULL on that catalog. Perhaps some backend sees the old relfilenode
value and tries to a heap scan, interpreting the now-truncated file as
empty?

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company