Hello Chris, one way in which you could avoid that problem is to use simfs instead of ploop for that container. Otherwise, yes, you should compact containers on a regular basis (preferably during off peak hours...).
HTH Best, Corrado ___________________________________________ > On 3 Oct 2016, at 11:26 PM, Chris James <cja...@wiredtree.com> wrote: > > Howdy, > > I'm running a graphing container on VZ7. Specifically it's running > carbon-cache with metrics coming in every minute for a total of 46G of > stats spread out over 193349 files. This disk space usage of this > containers ploop container is regularly blowing up. > > Today, I ran the following command: > > prl_disk_tool compact --hdd /vz/private/101/root.hdd > > This freed up 172GB of disk space. > > How do I prevent this kind of fragmentation from building up? Am I > expected to compact the ploop images on a regular basis? > > Thanks, > Chris > _______________________________________________ > Users mailing list > Users@openvz.org > https://lists.openvz.org/mailman/listinfo/users _______________________________________________ Users mailing list Users@openvz.org https://lists.openvz.org/mailman/listinfo/users
