Alvaro Herrera <alvherre@commandprompt.com> writes:
> Tom Lane escribi�:
>> The other problem with it is that it's running vacuum() in an
>> indefinite-lifespan context. Perhaps that has something to do with
>> the report we saw awhile back of autovac leaking memory ...
> Hmm, I'm not sure which memory leak are you referring to, but if it's
> the same I'm thinking of, then it cannot be the same because this one
> occurs on the worker and the other was on the launcher; also, I patched
> that one:
I was thinking of Erik Jones' report of TopMemoryContext bloat in a
database with 200000 tables (in pre-8.3 code). But I guess this still
doesn't fit, because any leakage induced by vacuum() would have been in
AutovacMemCxt, and that wasn't what he saw. So I still don't know what
was happening with Erik's issue.
regards, tom lane
