Re: Slow catchup of 2PC (twophase) transactions on replica in LR

Vitaly Davydov <v.davydov@postgrespro.ru>

From: Давыдов Виталий <v.davydov@postgrespro.ru>

To: pgsql-hackers@lists.postgresql.org

Date: 2024-02-29T17:34:42Z

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 →

Fix random failure in 021_twophase.
- 0dcea330babd 18.0 landed
Allow altering of two_phase option of a SUBSCRIPTION.
- 1462aad2e447 18.0 landed
Doc: use true|false rather than on|off for "failover" option
- fa65a022db26 17.0 cited
Support an optional asynchronous commit mode, in which we don't flush WAL
- 4a78cdeb6b59 8.3.0 cited

Attachments

0001-Async-commit-support-for-twophase-transactions.patch (text/x-patch) patch 0001

Dear All,

Consider, please, my patch for async commit for twophase transactions. It can be applicable when catchup performance is not enought with publication parameter twophase = on.

The key changes are:
 * Use XLogSetAsyncXactLSN instead of XLogFlush as it is for usual transactions. * In case of async commit only, save 2PC state in the pg_twophase file (but not fsync it) in addition to saving in the WAL. The file is used as an alternative to storing 2pc state in the memory. * On recovery, reject pg_twophase files with future xids.Probably, 2PC async commit should be enabled by a GUC (not implemented in the patch).

With best regards,
Vitaly