| From: | shihao zhong <zhong950419(at)gmail(dot)com> |
|---|---|
| To: | Michael Paquier <michael(at)paquier(dot)xyz> |
| Cc: | pgsql-hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>, Robert Haas <robertmhaas(at)gmail(dot)com> |
| Subject: | Re: Racing in ROLLBACK PREAPRED and checkpoint |
| Date: | 2026-11-02 03:37:16 |
| Message-ID: | CAGRkXqTWisELby9DPDczh_AT6birh-B=a-D8n0Tz-KJHxW4WLA@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hi Michael,
Thanks for looking into this.
I went through the other places that write a WAL record and then
change state races with the checkpointer. I found three more cases
and reported them in separate threads.
One is multixact creation [1].
The other two are CLOG truncation and hash index bucket allocation [2].
The other paths I checked are safe.
[1]
https://www.postgresql.org/message-id/CAGRkXqSuiJ_X4+Z6rsD42pcuACnD9PhoQ-Os+mUJNmJarBWkNg@mail.gmail.com
[2]
https://www.postgresql.org/message-id/CAGRkXqTSAb-H4oxk=J9QBv-Kh2-o=cnCWuiSduMVaVzrKYmKXQ@mail.gmail.com
Thanks,
Shihao
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Jonathan S. Katz | 2026-11-02 03:54:48 | PostgreSQL 19 GA press release draft |
| Previous Message | Zhijie Hou | 2026-11-02 02:51:25 | Re: Avoid resetting subscription two_phase state to pending when already enabled |