Re: Minimal logical decoding on standbys

Jeff Davis <pgsql@j-davis.com>

From: Jeff Davis <pgsql@j-davis.com>

To: "Drouvot, Bertrand" <bertranddrouvot.pg@gmail.com>, Andres Freund <andres@anarazel.de>

Cc: Robert Haas <robertmhaas@gmail.com>, Thomas Munro <thomas.munro@gmail.com>, Alvaro Herrera <alvherre@2ndquadrant.com>, Ibrar Ahmed <ibrar.ahmad@gmail.com>, Amit Khandekar <amitdkhan.pg@gmail.com>, fabriziomello@gmail.com, tushar <tushar.ahuja@enterprisedb.com>, Rahila Syed <rahila.syed@2ndquadrant.com>, pgsql-hackers <pgsql-hackers@postgresql.org>

Date: 2023-04-02T21:25:35Z

Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

Reduce the log level in 035_standby_logical_decoding.pl.
- 3034dc56ef4b 16.0 landed
035_standby_logical_decoding: Add missing waits for replication
- 57411c82ce86 16.0 landed
For cascading replication, wake physical and logical walsenders separately
- e101dfac3a53 16.0 landed
Handle logical slot conflicts on standby
- 26669757b6a7 16.0 landed
Support invalidating replication slots due to horizon and wal_level
- be87200efd93 16.0 landed
Prevent use of invalidated logical slot in CreateDecodingContext()
- 4397abd0a2af 16.0 landed
Replace replication slot's invalidated_at LSN with an enum
- 15f8203a5975 16.0 landed
Pass down table relation into more index relation functions
- 61b313e47eb9 16.0 landed
Assert only valid flag bits are passed to visibilitymap_set()
- a88a18b1250b 16.0 landed
Remove unused _bt_delitems_delete() argument.
- dc43492e46c7 14.0 cited
Add xl_btree_delete optimization.
- d2e5e20e5711 13.0 cited

On Fri, 2023-03-31 at 02:44 -0700, Jeff Davis wrote:
> Thank you, done. I think the nearby line was also wrong, returning
> true
> when there was no timeout. I combined the lines and got rid of the
> early return so it can check the list and timeout condition like
> normal. Attached.

On second (third?) thought, I think I was right the first time. It
passes the flag WL_EXIT_ON_PM_DEATH (included in the
ConditionVariableWaitSet), so a WL_POSTMASTER_DEATH event should not be
returned.

Also, I think the early return is correct. The current code in
ConditionVariableTimedSleep() still checks the wait list even if
WaitLatch() returns WL_TIMEOUT (it ignores the return), but I don't see
why it can't early return true. For a socket event in
ConditionVariableEventSleep() I think it should early return false.

Regards,
	Jeff Davis