Horton

RE: Skipping logical replication transactions on subscriber side

Zhijie Hou (Fujitsu) <houzj.fnst@fujitsu.com>

View thread View on postgr.es
From: "houzj.fnst@fujitsu.com" <houzj.fnst@fujitsu.com>
To: Masahiko Sawada <sawada.mshk@gmail.com>
Cc: Amit Kapila <amit.kapila16@gmail.com>, "tanghy.fnst@fujitsu.com" <tanghy.fnst@fujitsu.com>, "osumi.takamichi@fujitsu.com" <osumi.takamichi@fujitsu.com>, Alexey Lesovsky <lesovsky@gmail.com>, Peter Eisentraut <peter.eisentraut@enterprisedb.com>, PostgreSQL-development <pgsql-hackers@postgresql.org>, Greg Nancarrow <gregn4422@gmail.com>
Date: 2021-10-20T03:02:55Z
Lists: pgsql-hackers 
On Mon, Oct 18, 2021 9:34 AM Masahiko Sawada <sawada.mshk@gmail.com> wrote:
> I've attached updated patches that incorporate all comments I got so far.

Hi,

Here are some minor comments for the patches.

v17-0001-Add-a-subscription-errors-statistics-view-pg_sta.patch

1)

+	/* Clean up */
+	if (not_ready_rels != NIL)
+		list_free_deep(not_ready_rels);

Maybe we don't need the ' if (not_ready_rels != NIL)' check as
list_free_deep will do this check internally.

2)

+	for (int i = 0; i < msg->m_nentries; i++)
+	{
+		HASH_SEQ_STATUS sstat;
+		PgStat_StatSubWorkerEntry *wentry;
+
+		/* Remove all worker statistics of the subscription */
+		hash_seq_init(&sstat, subWorkerStatHash);
+		while ((wentry = (PgStat_StatSubWorkerEntry *) hash_seq_search(&sstat)) != NULL)
+		{
+			if (wentry->key.subid == msg->m_subids[i])
+				(void) hash_search(subWorkerStatHash, (void *) &(wentry->key),
+								   HASH_REMOVE, NULL);

Would it be a little faster if we scan hashtable in outerloop and
scan the msg in innerloop ?
Like:
while ((wentry = (PgStat_StatSubWorkerEntry *) hash_seq_search(&sstat)) != NULL)
{
	for (int i = 0; i < msg->m_nentries; i++)
	...


v17-0002-Add-RESET-command-to-ALTER-SUBSCRIPTION-command

1)
I noticed that we cannot RESET slot_name while we can SET it.
And the slot_name have a default behavior that " use the name of the subscription for the slot name.".
So, is it possible to support RESET it ?

Best regards,
Hou zj

Commits

  1. Test ALIGNOF_DOUBLE==4 compatibility under ALIGNOF_DOUBLE==8.

  2. Reorder subskiplsn in pg_subscription to avoid alignment issues.

  3. Add ALTER SUBSCRIPTION ... SKIP.

  4. Optionally disable subscriptions on error.

  5. Update docs of logical replication for commit 8d74fc96db.

  6. Respect permissions within logical replication.

  7. Fix regression test failure caused by commit 8d74fc96db.

  8. Add a view to show the stats of subscription workers.

  9. Add logical change details to logical replication worker errcontext.

  10. Rename LOGICAL_REP_MSG_STREAM_END to LOGICAL_REP_MSG_STREAM_STOP.

  11. Fix typo in protocol.sgml.

  12. Remove unused argument in apply_handle_commit_internal().

  13. Fix replication of in-progress transactions in tablesync worker.

  14. Reorder pg_sequence columns to avoid alignment issue