| From: | Zhijie Hou <houzhijie22(at)gmail(dot)com> |
|---|---|
| To: | Bertrand Drouvot <bertranddrouvot(dot)pg(at)gmail(dot)com> |
| Cc: | Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com>, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Assertion failure in update_local_synced_slot(), slotsync.c:364 |
| Date: | 2026-11-17 09:27:24 |
| Message-ID: | b8b71aca-e226-4c1e-adb4-3a5faa34acb6@gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hi,
On Tue, Nov17, 2026 at 04:45 PM Bertrand Drouvot wrote:
>
> On Mon, Nov 16, 2026 at 04:09:44PM +0800, Zhijie Hou wrote:
> >
> > Attached is the v3 patch set (with the TAP test split into 0002).
> > Please let me know what you think.
>
> Thanks for the new patch!
>
> I've one nit comment:
>
> + if (slot->data.persistency == RS_TEMPORARY && !found_consistent_snapshot)
> + {
> + ereport(LOG,
>
> When no attempt is made to build a consistent snapshot, this could
> still be emitted on every sync cycle. So I wonder if DEBUG1 wouldn't
> be enough here?
Thanks for the comments.
I considered this before but left it unchanged as I thought the
frequency of such logging would be low because 1) this only applies to
newly synced slots, 2) slotsync increases the interval for the next
cycle when the slot on the primary isn’t advancing. Also, other skip
reason is also using LOG level, so wanted to be consistent.
--
Best Regards,
Zhijie Hou
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Jingtang Zhang | 2026-11-17 09:35:48 | Re: transaction_timeout is not enforced during ClientWrite |
| Previous Message | Koshino Taiki | 2026-11-17 08:58:51 | Re: Proposal: Include CHECK constraint expression in violation error DETAIL |