public inbox for [email protected]
help / color / mirror / Atom feedFrom: Imseih (AWS), Sami <[email protected]>
To: Kyotaro Horiguchi <[email protected]>
To: [email protected] <[email protected]>
Cc: [email protected] <[email protected]>
Subject: Re: [BUG] Panic due to incorrect missingContrecPtr after promotion
Date: Fri, 27 May 2022 02:01:27 +0000
Message-ID: <[email protected]> (raw)
In-Reply-To: <[email protected]>
References: <[email protected]>
<[email protected]>
<[email protected]>
<[email protected]>
<[email protected]>
<[email protected]>
<[email protected]>
<[email protected]>
After further research, we found the following.
Testing on 13.6 with the attached patch we see
that the missingContrecPtr is being incorrectly
set on the standby and the promote in the tap
test fails.
Per the comments in xlog.c, the
missingContrecPtr should not be set when
in standby mode.
/*
* When not in standby mode we find that WAL ends in an incomplete
* record, keep track of that record. After recovery is done,
* we'll write a record to indicate downstream WAL readers that
* that portion is to be ignored.
*/
if (!StandbyMode &&
!XLogRecPtrIsInvalid(xlogreader->abortedRecPtr))
{
abortedRecPtr = xlogreader->abortedRecPtr;
missingContrecPtr = xlogreader->missingContrecPtr;
elog(LOG, "missingContrecPtr == %ld", missingContrecPtr);
}
If StandbyModeRequested is checked instead, which
checks for the presence of a standby signal file,
The missingContrecPtr is not set on the
standby and the test succeeds.
if (!StandbyModeRequested &&
!XLogRecPtrIsInvalid(xlogreader->abortedRecPtr))
{
abortedRecPtr = xlogreader->abortedRecPtr;
missingContrecPtr = xlogreader->missingContrecPtr;
elog(LOG, "missingContrecPtr == %ld", missingContrecPtr);
}
If this is a bug as it appears, it appears the original patch
to resolve this issue is not needed and the ideal fix
Is to ensure that a standby does not set
missingContrecPtr.
Would like to see what others think.
Thanks
--
Sami Imseih
Amazon Web Services
Attachments:
[application/octet-stream] 0001-Testing_patch[2][2].patch (1.5K, ../[email protected]/2-0001-Testing_patch%5B2%5D%5B2%5D.patch)
download | inline diff:
From 736308047747c7b1e5b78dacf95a0b8459d2b7a1 Mon Sep 17 00:00:00 2001
From: "Imseih (AWS)" <[email protected]>
Date: Thu, 26 May 2022 19:41:22 -0500
Subject: [PATCH 1/1] Testing_patch
---
src/backend/access/transam/xlog.c | 2 ++
src/test/recovery/t/026_overwrite_contrecord.pl | 2 +-
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/src/backend/access/transam/xlog.c b/src/backend/access/transam/xlog.c
index 3d76fad..a1b78f3 100644
--- a/src/backend/access/transam/xlog.c
+++ b/src/backend/access/transam/xlog.c
@@ -4381,6 +4381,7 @@ ReadRecord(XLogReaderState *xlogreader, int emode,
{
abortedRecPtr = xlogreader->abortedRecPtr;
missingContrecPtr = xlogreader->missingContrecPtr;
+ elog(LOG, "missingContrecPtr == %ld", missingContrecPtr);
}
if (readFile >= 0)
@@ -4398,6 +4399,7 @@ ReadRecord(XLogReaderState *xlogreader, int emode,
if (errormsg)
ereport(emode_for_corrupt_record(emode, EndRecPtr),
(errmsg_internal("%s", errormsg) /* already translated */ ));
+
}
/*
diff --git a/src/test/recovery/t/026_overwrite_contrecord.pl b/src/test/recovery/t/026_overwrite_contrecord.pl
index cb7f5b6..a3000f1 100644
--- a/src/test/recovery/t/026_overwrite_contrecord.pl
+++ b/src/test/recovery/t/026_overwrite_contrecord.pl
@@ -101,6 +101,6 @@ like(
$log,
qr[successfully skipped missing contrecord at],
"found log line in standby");
-
+$node_standby->promote;
$node->stop;
$node_standby->stop;
--
2.32.1 (Apple Git-133)
view thread (34+ messages) latest in thread
reply
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: [email protected]
Cc: [email protected], [email protected], [email protected]
Subject: Re: [BUG] Panic due to incorrect missingContrecPtr after promotion
In-Reply-To: <[email protected]>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox