Horton

Re: pg_restore --no-policies should not restore policies' comment

Alexander Lakhin <exclusion@gmail.com>

View thread
From: Alexander Lakhin <exclusion@gmail.com>
To: jian he <jian.universality@gmail.com>, Fujii Masao <masao.fujii@gmail.com>
Cc: Fujii Masao <masao.fujii@oss.nttdata.com>, PostgreSQL-development <pgsql-hackers@postgresql.org>
Date: 2025-10-17T05:00:00Z
Lists: pgsql-hackers 
Hello Jian,

16.10.2025 17:22, jian he пишет:
>>> Buildfarm member petalura reported a failure at [1] after 0001 was backpatched
>>> to v13. However, the server logs show a "double free or corruption" error
>>> occurring as the backend was exiting. Since that code was not touched by 0001,
>>> for now I don't think this failure is related to the patch...
>>> But am I missing something?
>> After pushing 0004, petalura reported another failure with a "double
>> free or corruption" error:
>> https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=petalura&dt=2025-09-18%2002%3A14%3A32
>>
>> Buildfarm member phycodurus also reported similar failures:
>> https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=phycodurus&dt=2025-09-16%2011%3A09%3A07
>> https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=phycodurus&dt=2025-09-16%2016%3A44%3A37
>>
>> I'll dig deeper into this issue.
>>
> It appears that this issue no longer exists, perhttps://bfbot.cputube.org

Unfortunately, buildfarm animals (especially phycodurus) are still
failing due to "double free or corruption (!prev)":
https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=phycodurus&dt=2025-10-16%2019%3A07%3A52

I can try to reproduce this locally if it can be helpful.

Best regards,
Alexander