On Wed, Mar 25, 2020 at 12:27:56PM +0100, Eneko Lacunza wrote: > Hi Alwin, > > El 25/3/20 a las 11:55, Alwin Antreich escribió: > > > > > > > > The easiest way ist to destroy and re-create the OSD with a bigger > > > > > > DB/WAL. The guideline from Facebook for RocksDB is 3/30/300 GB. > > > > > It's well below the 3GiB limit in the guideline ;) > > > > For now. ;) > > > Cluster has 2 years now, data amount is quite stable, I think it will hold > > > for some time ;) > > Hm... Igor recons that this seems to be normal. > > https://tracker.ceph.com/issues/38745#note-28 > Thanks for the heads up, I didn't receive an email from tracker about the > update. Just replied with requested info. > > I think it's "normal" in the sense that it is the way code works now, but > there doesn't seem to be a good reason. I'm not really worried about the > performance, but about having to suppress the spillover warning ;) Yeah. I hope this has been resolved when we introduce Ceph Octopus. :)
-- Cheers, Alwin _______________________________________________ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user