I can't say for sure, but my thoughts: - It probably has nothing to do with pacemaker, cluster and drbd - SQL databases use garbage collector, which could do one big transaction on restart and freed some space - Filesystem itself may be using some sort of garbage collector, especially on remounting. For example in btrfs there are delays between file deletions and reporting space freed.
2016-10-04 13:02 GMT+03:00 Ulrich Windl <[email protected]>: > >>> Eric Robinson <[email protected]> schrieb am 04.10.2016 um > 09:24 in > Nachricht > <DM5PR03MB2729E0F0799541985B326DF7FAC50@DM5PR03MB2729. > namprd03.prod.outlook.com> > > > The filesystem on my corosync+pacemaker cluster is 1TiB in size and was > 95% > > full, with only 54GB available. Drbd was UpToDate/UpToDate. > > > > I restarted Pacemaker, and after that my filesystem now shows 49% full > with > > 300GB+ free space. > > > > I checked and there does not seem to be any data missing. All MySQL > > databases are up to date. > > > > Can anyone think of a reason that the filesystem numbers would change so > > dramatically when all I did was restart Pacemaker? > > An open file had been "removed", so that it was actually removed when the > process died? > > > > > -- > > Eric Robinson > > > > > > _______________________________________________ > > Users mailing list: [email protected] > > http://clusterlabs.org/mailman/listinfo/users > > > > Project Home: http://www.clusterlabs.org > > Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf > > Bugs: http://bugs.clusterlabs.org > > > > > > _______________________________________________ > Users mailing list: [email protected] > http://clusterlabs.org/mailman/listinfo/users > > Project Home: http://www.clusterlabs.org > Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf > Bugs: http://bugs.clusterlabs.org >
_______________________________________________ Users mailing list: [email protected] http://clusterlabs.org/mailman/listinfo/users Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf Bugs: http://bugs.clusterlabs.org
