From: Alvaro Herrera Date: Wed, 23 Mar 2022 17:22:10 +0000 (+0100) Subject: Fix "missing continuation record" after standby promotion X-Git-Tag: REL_10_21~41 X-Git-Url: http://git.postgresql.org/gitweb/?a=commitdiff_plain;h=9ade3c09a5d11f4be91b5e180832abd2dbc2c55d;p=postgresql.git Fix "missing continuation record" after standby promotion Invalidate abortedRecPtr and missingContrecPtr after a missing continuation record is successfully skipped on a standby. This fixes a PANIC caused when a recently promoted standby attempts to write an OVERWRITE_RECORD with an LSN of the previously read aborted record. Backpatch to 10 (all stable versions). Author: Sami Imseih Reviewed-by: Kyotaro Horiguchi Reviewed-by: Álvaro Herrera Discussion: https://postgr.es/m/44D259DE-7542-49C4-8A52-2AB01534DCA9@amazon.com --- diff --git a/src/backend/access/transam/xlog.c b/src/backend/access/transam/xlog.c index 8e8bdde7646..6772e248222 100644 --- a/src/backend/access/transam/xlog.c +++ b/src/backend/access/transam/xlog.c @@ -10261,6 +10261,10 @@ VerifyOverwriteContrecord(xl_overwrite_contrecord *xlrec, XLogReaderState *state (uint32) (state->overwrittenRecPtr >> 32), (uint32) state->overwrittenRecPtr); + /* We have safely skipped the aborted record */ + abortedRecPtr = InvalidXLogRecPtr; + missingContrecPtr = InvalidXLogRecPtr; + ereport(LOG, (errmsg("successfully skipped missing contrecord at %X/%X, overwritten at %s", (uint32) (xlrec->overwritten_lsn >> 32), diff --git a/src/test/recovery/t/026_overwrite_contrecord.pl b/src/test/recovery/t/026_overwrite_contrecord.pl index 57b2a6b7fb9..50902e59a56 100644 --- a/src/test/recovery/t/026_overwrite_contrecord.pl +++ b/src/test/recovery/t/026_overwrite_contrecord.pl @@ -15,7 +15,7 @@ plan tests => 3; # Test: Create a physical replica that's missing the last WAL file, # then restart the primary to create a divergent WAL file and observe # that the replica replays the "overwrite contrecord" from that new -# file. +# file and the standby promotes successfully. my $node = PostgresNode->get_new_node('primary'); $node->init(allows_streaming => 1); @@ -105,5 +105,8 @@ like( qr[successfully skipped missing contrecord at], "found log line in standby"); +# Verify promotion is successful +$node_standby->promote; + $node->stop; $node_standby->stop;