Re: [HACKERS] Initializing initFileRelationIds list via write is unsafe

2015-06-08 Thread Robert Haas
On Sat, Jun 6, 2015 at 11:51 PM, Tom Lane wrote: > This suggests that CLOBBER_CACHE_ALWAYS is actually missing a pretty > large part of the cache behavioral space. Maybe we should devise some > sort of "CLOBBER_CACHE_RANDOMLY" option that would inject cache flush > events more selectively, perhap

Re: [HACKERS] Initializing initFileRelationIds list via write is unsafe

2015-06-06 Thread Tom Lane
I wrote: >> I've been chasing the intermittent "cache lookup failed for access method >> 403" failure at session startup that's been seen lately in the buildfarm, >> for instance here: >> http://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=axolotl&dt=2015-06-04%2019%3A22%3A46 >> (Axolotl has sho

[HACKERS] Initializing initFileRelationIds list via write is unsafe

2015-06-06 Thread Tom Lane
I wrote: > I've been chasing the intermittent "cache lookup failed for access method > 403" failure at session startup that's been seen lately in the buildfarm, > for instance here: > http://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=axolotl&dt=2015-06-04%2019%3A22%3A46 > (Axolotl has shown th