| From: | Kyotaro Horiguchi <horikyota(dot)ntt(at)gmail(dot)com> |
|---|---|
| To: | bharath(dot)rupireddyforpostgres(at)gmail(dot)com |
| Cc: | nathandbossart(at)gmail(dot)com, cary(dot)huang(at)highgo(dot)ca, pgsql-hackers(at)lists(dot)postgresql(dot)org, satyanarlapuram(at)gmail(dot)com |
| Subject: | Re: Switching XLog source from archive to streaming when primary available |
| Date: | 2022-09-09 05:16:17 |
| Message-ID: | 20220909.141617.1843398696475923032.horikyota.ntt@gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Being late for the party.
It seems to me that the function is getting too long. I think we
might want to move the core part of the patch into another function.
I think it might be better if intentionalSourceSwitch doesn't need
lastSourceFailed set. It would look like this:
> if (lastSourceFailed || switchSource)
> {
> if (nonblocking && lastSourceFailed)
> return XLREAD_WOULDBLOCK;
+ if (first_time)
+ last_switch_time = curr_time;
..
+ if (!first_time &&
+ TimestampDifferenceExceeds(last_switch_time, curr_time,
..
+ /* We're not here for the first time any more */
+ if (first_time)
+ first_time = false;
I don't think the flag first_time is needed.
regards.
--
Kyotaro Horiguchi
NTT Open Source Software Center
| From | Date | Subject | |
|---|---|---|---|
| Next Message | John Naylor | 2022-09-09 05:18:20 | Re: build remaining Flex files standalone |
| Previous Message | houzj.fnst@fujitsu.com | 2022-09-09 05:11:03 | When should we bump the logical replication protocol version? |