Re: SIGSEGV in BRIN autosummarize

Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org> writes:
> And the previous code crashes in 45 minutes? That's solid enough for
> me; I'll clean up the patch and push in the next few days. I think what
> you have now should be sufficient for the time being for your production
> system.

I'm still of the opinion that the presented patch isn't preventing any
crashes. The failure to check avw_database would generally lead to
tasks silently getting dropped in the mistaken belief that the target
table has gone away. I could believe that a crash occurs if the given
schema OID exists in some other DB, and the given table OID does too,
but it's for a table/index of the wrong kind ... but how likely is that?
Not very likely at all, given the way we generate OIDs.

(And if it did happen, I'd say the cause is a failure to recheck relkind
somewhere downstream of the autovac task dispatch code, anyway.)

regards, tom lane