| From: | Andreas Karlsson <andreas(at)proxel(dot)se> |
|---|---|
| To: | Ankush Mondal <ankushmondal1y2t(at)gmail(dot)com>, pgsql-hackers(at)lists(dot)postgresql(dot)org |
| Subject: | Re: [PATCH] standby: Clarify recovery conflict comment |
| Date: | 2026-11-13 08:45:34 |
| Message-ID: | c5f767fd-12a1-4581-a3d9-e03b5bcd1f17@proxel.se |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On 11/13/26 09:19, Ankush Mondal wrote:
> if (GetCurrentTimestamp() >= ltime && ltime != 0)
> {
> /*
> - * We're already behind, so clear a path as quickly as possible.
> + * The standby delay has expired, so signal conflicting backends
> + * to release their buffer pins as quickly as possible.
> */
> SendRecoveryConflictWithBufferPin(RECOVERY_CONFLICT_BUFFERPIN);
> }
Hi,
Thanks for the patch but I do not think the new version is better. The
old version explained intent while the new comment just re-explains what
the code does. Maybe some mix of the two comments would be better but I
doubt it is worth the churn.
--
Andreas Karlsson
Percona
| From | Date | Subject | |
|---|---|---|---|
| Next Message | 신성준 | 2026-11-13 08:48:10 | Re: Add wait events for server logging destination writes |
| Previous Message | Jingtang Zhang | 2026-11-13 08:44:29 | transaction_timeout is not enforced during ClientWrite |