Re: Crash report for some ICU-52 (debian8) COLLATE and work_mem values

Daniel Verite <daniel@manitou-mail.org>

From: "Daniel Verite" <daniel@manitou-mail.org>
To: "Peter Eisentraut" <peter.eisentraut@2ndquadrant.com>
Cc: "Peter Geoghegan" <pg@bowt.ie>,"Tom Lane" <tgl@sss.pgh.pa.us>,"PostgreSQL mailing lists" <pgsql-bugs@postgresql.org>
Date: 2017-08-07T17:39:38Z
Lists: pgsql-bugs, pgsql-hackers
	Peter Eisentraut wrote:

> In my testing, there are no crashes in ICU 51 and ICU 53, so we could
> restrict this workaround to version 52 specifically.  Perhaps someone
> else can confirm my test results.

I'm testing all collations with ICU-51.2 with the same data as previously,
and I just see it crashing at the 103th and a few other az-* after that:

2017-08-07 19:09:26.538 CEST [59353] LOG:  server process (PID 60458) was
terminated by signal 11: Segmentation fault
2017-08-07 19:09:26.538 CEST [59353] DETAIL:  Failed process was running:
select count(distinct wordtext collate "az-Latn-AZ-u-co-search-x-icu") from
words_test
2017-08-07 19:09:26.538 CEST [59353] LOG:  terminating any other active
server processes
2017-08-07 19:09:26.538 CEST [60316] WARNING:  terminating connection because
of crash of another server process

I don't have core files for this run but that's probably enough to say
that ICU-51 is no good for us.

Best regards,
-- 
Daniel Vérité
PostgreSQL-powered mailer: http://www.manitou-mail.org
Twitter: @DanielVerite


Commits

  1. Reject use of ucol_strcollUTF8() before ICU 53

  2. Rethink behavior of pg_import_system_collations().

  3. Faster expression evaluation and targetlist projection.