Search results

PBS has a 'verify' of the datastore. How often are we supposed to run that? Daily? Weekly?

Correct. We have a 10Gbit connection between the two datacenters. Although dadup is spinning disk, there are about 70 of them, so combined (Ceph) the cluster is able to perform pretty well. Latency between the VPS and daDup is about 0.5ms. When not in beta anymore, I think we would put the PBS...

Within the European Economic Zone, yes. Outside that, the hassle with VAT is too much. Feel free to request the beta, I'll set you up this afternoon.

I created bug https://bugzilla.proxmox.com/show_bug.cgi?id=3047 for looking into performance, so I'm diving into that. I think you can only grant access to /remote, which would allow every user to see everyones remotes, that's not very cool. I created feature-request...

Yes, that would probably be the target-price. Probably not with the free 500GB. Yes, both would be perfect. There is currently 1TB of backups in the pool. I don't think a cache disk will help much (it's there now, but not much reading from it).

As with most commercial companies, we try to make money, so yes. Everybody that helped get the Proxmox product stable has also been able to use free backupservices for a few months. If they feel it's worth their while is up to them.

Yes. This is an issue. Although backups and other tasks seems to be running fine, the pvestatd-calls hallt somewhere. Storage is not too fast, but metadata is on a 'special' device (and that one is fast). I also added a log-device to the zpool yesterday, which doesn't seem to be used. I'll add...

Yes.

I'm trying to find a maxthreads or maxconnections value in the code, but I'm failing to. @fabian or @Thomas Lamprecht , could it be a MaxClients-kinda limit?

Request timed out on localhost:8007 .

No. My monitoring runs locally and complained around the same time.

Yes, that is indeed related. It does seem that PBS was unreachable, but the weird thing is that the server is not complaining about anything...

Correct. I got monitoring notifications around that time. But the PBS was pretty much idle then.

Yes, the singlethreadedness of pvestatd is a problem that causes lots of confusion. I'll create a PVE-bug for that. As far as PBS is concerned, I see no signs of issues on that side. Logging continues as normal and statistics do not show storage or CPU to be very busy. Performance of the PBS...

The pvestatd is waiting for responses from PBS which was pretty slow earlier. We just tuned the ZFS filesystem a bit with a 'special' device and upgraded to 0.8-19 which seems to have improved webserver-responses a bit.

You could add both datastores and their encryption keys..

Correct! The 500GB is per user. Let me know what second name you want to use for the second cluster.

Feel free to reach out through the Forum as well!

After doing an encrypted backup, when I try to restore I get: proxmox-backup-client failed: Error: wrong signature in manifest (500) PBS confirms that encryption is 'mixed', but the disk and config are encrypted. Am I doing it wrong? :)

Probably, one of the OSD's restarted. While restarting, CRUSH remapped the PG's on that OSD. I recently learned that this remapping sometimes failes to often (more than 5 times, by default) and CRUSH gives up. Find out which OSD restarted, and try to play around with the crush weight of that...