Correction; The original patch on the bug report seems quite sufficient for the branch targeted. A PR with MDEV-35904 would be much appreciated.
On Tue, 10 Jun 2025 at 17:51, Daniel Black <dan...@mariadb.org> wrote: > > > On Mon, 9 Jun 2025 at 16:48, Otto Kekäläinen via discuss < > discuss@lists.mariadb.org> wrote: > >> > Patching one of our spare servers this morning and ran into a couple of >> issues. The first was a complaint in the service status: >> > >> > mariadb.service: Referenced but unset environment variable evaluates to >> an empty string: MYSQLD_OPTS, _WSREP_NEW_CLUSTER >> > >> > Which was fixed by running systemctl edit mariadb.service and adding >> > >> > Environment="MYSQLD_OPTS=" >> > Environment="_WSREP_NEW_CLUSTER=" >> >> This is just a warning, and does not affect anything. Naturally it >> should be fixed though. Potential fix posted in >> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1105029 >> >> > A more complete backport of MDEV-19210 commits would be required to have a > serviceable Galera with systemd. > > upstream commits: > * 3d2e54ff8c6070800ead60dd1616df30c5936c8b > * 105473233dd49b198758432fa0da5d0811f99c08 > * aeffec60f6864bae5af04dac1184f2a0f2c77f38 > * 4c56c66372c3f578dc75b537a79bf922dbddcafd > * 621926e90ac05c33b956cf123f08e56b654d00df > * 7936254012784790b7747dc07941917af1bbb83c > > > >> > The second was from the error log which stated that the service was >> unable to read /etc/mysql/rest/keyfile.passwd, which turned out to be a >> permissions error on /etc/mysql/rest which was 600 and not 700 (I checked >> the other servers and theirs were also 600 so I guess it's the fix for >> MDEV-36229 which caused that) >> > > yes, directories should have execute permissions. > >
_______________________________________________ discuss mailing list -- discuss@lists.mariadb.org To unsubscribe send an email to discuss-le...@lists.mariadb.org
