Re: subscriptionCheck failures on nightjar

Michael Paquier <michael@paquier.xyz>

From: Michael Paquier <michael@paquier.xyz>
To: Tomas Vondra <tomas.vondra@2ndquadrant.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Robert Haas <robertmhaas@gmail.com>, Andres Freund <andres@anarazel.de>, Thomas Munro <thomas.munro@enterprisedb.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2019-09-18T00:58:15Z
Lists: pgsql-hackers
On Tue, Sep 17, 2019 at 09:45:10PM +0200, Tomas Vondra wrote:
> FWIW I agree with Andres that there probably is an actual bug. The file
> should not just disappear like this, it's clearly unexpected so the
> PANIC does not seem entirely inappropriate.

Agreed.

> I've tried reproducing the  issue on my local systems, with the extra
> sleeps between fsyncs and so on, but I haven't managed to trigger it so
> far :-(

On my side, I have let this thing run for a couple of hours with a
patched version to include a sleep between the rename and the sync but
I could not reproduce it either:
#!/bin/bash
attempt=0
while true; do
	attempt=$((attempt+1))
	echo "Attempt $attempt"
	cd $HOME/postgres/src/test/recovery/
	PROVE_TESTS=t/006_logical_decoding.pl make check > /dev/null 2>&1
	ERRNUM=$?
	if [ $ERRNUM != 0 ]; then
		echo "Failed at attempt $attempt"
		exit $ERRNUM
	fi
done
> Yes, it should be moved to the older section - it's clearly a v11 bug.

And agreed.
--
Michael

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →
  1. Fix "base" snapshot handling in logical decoding