| From: | Andres Freund <andres(at)anarazel(dot)de> |
|---|---|
| To: | Michael Paquier <michael(at)paquier(dot)xyz> |
| Cc: | Alvaro Herrera <alvherre(at)2ndquadrant(dot)com>, Kyotaro HORIGUCHI <horiguchi(dot)kyotaro(at)lab(dot)ntt(dot)co(dot)jp>, pgsql-hackers(at)postgresql(dot)org, petr(at)2ndquadrant(dot)com, simon(at)2ndQuadrant(dot)com |
| Subject: | Re: Non-reserved replication slots and slot advancing |
| Date: | 2018-07-10 07:26:30 |
| Message-ID: | 20180710072630.oosunvidrxjdjdbb@alap3.anarazel.de |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hi,
On 2018-07-10 09:54:28 +0900, Michael Paquier wrote:
> > > Also, ERRCODE_FEATURE_NOT_SUPPORTED doesn't quite seem right. How about
> > > ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE?
> >
> > +1 to both of Andres' suggestions.
>
> Those indeed sound better. What do you think of the attached?
Looks generally good. One remark:
> + /* A slot whose restart_lsn has never been reserved cannot be advanced */
> + if (XLogRecPtrIsInvalid(MyReplicationSlot->data.restart_lsn))
> + {
> + ReplicationSlotRelease();
> + ereport(ERROR,
> + (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
> + errmsg("cannot advance replication slot that has not previously reserved WAL")));
> + }
Why is the ReplicationSlotRelease() needed here? Souldn't the error
handling code do so automatically?
Greetings,
Andres Freund
| From | Date | Subject | |
|---|---|---|---|
| Next Message | 杨杰 | 2018-07-10 07:41:06 | user-friendliness improvement of pageinspect |
| Previous Message | Andres Freund | 2018-07-10 07:14:45 | Re: Recovery performance of standby for multiple concurrent truncates on large tables |