Horton

Re: Intermittent buildfarm failures on wrasse

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Andres Freund <andres@anarazel.de>
Cc: Peter Geoghegan <pg@bowt.ie>, Noah Misch <noah@leadboat.com>, David Rowley <dgrowleyml@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2022-04-15T05:18:59Z
Lists: pgsql-hackers 
Andres Freund <andres@anarazel.de> writes:
> On 2022-04-14 23:56:15 -0400, Tom Lane wrote:
>> Well, damn.  I changed my script that way and it failed on the tenth
>> iteration (versus a couple hundred successful iterations the other
>> way).

> Just to make sure: This is also on wrasse?

Right, gcc211 with a moderately close approximation to wrasse's
build details.  Why that shows the problem when we've not seen
it elsewhere remains to be seen.

> What DSM backend do we end up with on solaris? With shared memory stats
> we're using DSM a lot earlier and more commonly than before.

That ... is an interesting point.  It seems to be just "posix" though.

>> So somehow this is related to time-since-initdb, not
>> time-since-postmaster-start.  Any ideas?

> Perhaps it makes a difference that we start with a "young" database xid
> age wise?  We've had bugs around subtracting xids and ending up on some
> special one in the past.

It does seem like it's got to be related to small XID and/or small
LSN values.  No clue right now, but more news tomorrow, I hope.

			regards, tom lane

Commits

  1. Repurpose PROC_COPYABLE_FLAGS as PROC_XMIN_FLAGS

  2. Tighten ComputeXidHorizons' handling of walsenders.

  3. Adjust VACUUM's removable cutoff log message.

  4. Temporarily add some probes of tenk1's relallvisible in create_index.sql.

  5. Set synchronous_commit=on in test_setup.sql.

  6. Rearrange core regression tests to reduce cross-script dependencies.

  7. snapshot scalability: Don't compute global horizons while building snapshots.