Re: Too-long socket paths are breaking several buildfarm members

2022-07-03 Thread Michael Paquier
On Sun, Jul 03, 2022 at 09:40:23PM -0400, Tom Lane wrote: > Yeah, I just came to the same conclusion and pushed an equivalent > patch. Sorry for the duplicated effort. No problem. Thanks for the quick fix. -- Michael signature.asc Description: PGP signature

Re: Too-long socket paths are breaking several buildfarm members

2022-07-03 Thread Tom Lane
Michael Paquier writes: > There is PostgreSQL::Test::Utils::tempdir_short for that, which is > what all the nodes created in Cluster.pm use for > unix_socket_directories. One way to address the issue would be to > pass that to pg_upgrade with --socketdir, as of the attached. Yeah, I just came to

Re: Too-long socket paths are breaking several buildfarm members

2022-07-03 Thread Michael Paquier
On Sun, Jul 03, 2022 at 07:22:11PM -0400, Tom Lane wrote: > That path name is 3 bytes over the platform limit. Evidently, > "REL_15_STABLE" is just enough longer than "HEAD" to make this fail, > whereas we didn't see the problem as long as the test case only > ran in HEAD. That tells enough about