Search results

Please keep the tone civil here, we discourage everyone from passive-aggressive comments or disparaging speech. Especially as this creates a hostile atmosphere for newcomers. Thanks!

In short: more testing. Packages in the enterprise repositories are the same as in the non-subscription one, but they are tested by exposing them in the non-subscription repo first. So bugs are more likely to be caught before they can reach you. This leads to a more stable environment and...

If the disk remains readable, you can just copy the data over. Make sure to copy all folders from the datastore folder including the hidden .chunks folder. After you copied over all the data you can add the new disk as a datastore and run a verify job to make sure all the copied backups are...

Please calm down. Some things seem to have come across more offensive to you, than they read to me. Just to be clear: When we say “is not recommended” problems like the ones you ran into are to be expected. We “don't recommend” things, when we think they are just a flat-out bad idea, which are...

Hi, Not sure what you mean by that. There is no OpenZFS version 3.0 yet, at all [1]. Do you mean ZFS 2.3? [1]: https://github.com/openzfs/zfs/releases?q=3.0&expanded=true

I am guessing you are referring to this: [1]? Can't say I've seen this system deployed, but it looks like a fairly standard AMD Epyc system, so I don't see why this wouldn't be compatible with Proxmox VE or other products based on our Linux kernel. [1]...

Depends on how comfortable you are with using the CLI. In theory it shouldn't be to hard to use pvesh [1] and its json output format to get all the required information. From there you could writer a small script that parses the json and create a CSV file with that can be imported into Excel...

Hi, yes late loading is disabled. We base our kernel on the Ubuntu kernel and that feature is disabled there too [1]. And if you take a look at the thread you linked, you can also find out why [2]. So really, you don't want to "late load" your microcode. With that feature disabled, you'll need...

By the way this change was pushed to pbstest as PBS 3.4.1 by now, might want to check that out and see if it fixes the problem for you.

Well you did not define any keep options. In that case all backups will be kept, as a prune job that removes everything would be dangerous. This indicates that some backups are going to be removed after the grace period passes [1]. So they have not been marked as unused long enough to actually...

Did you set up a garbage collection job [1]? [1]: https://pbs.proxmox.com/docs/backup-client.html#garbage-collection

Yeah I only send the patch today, will probably take a little bit for this to get pushed out. You could space the push sync jobs more if that is an option for you. The problem here is that push sync jobs shouldn't take exclusive locks, as they only need to read the backup, not modify it...

Hi! two things: backup2 is the name of the datastore on the PBS and won't show up in the PVE storage menu at all. You can check which datastore is configured for the pbs-radxa storage by editing it. Are there any Sync Jobs configured on the PBS? If so, could you please tell us how they are...

Bitte gern und, naja, meist ist die Dokumentation schon ein bisschen hilfreicher als statistisch vermutlich korrekte Antworten ;)

Ok, dann hab ich das hier "immer nachts um 04:15 Uhr ausgeführt werden soll, als auch immer jeden 2ten Tag wie gehabt." falsch verstanden ;) Also es soll jeden zweiten Tag um 4:15 laufen? Dann ist das hier das einfachste: *-*-01..31/2 04:15 Das läuft dann an allen ungeraden tagen um 04:15. Ist...

Das sinnvollste wäre einfach zwei Backup Jobs zu machen. Einer der jeden Tag um 4:15 laüft und einer der zusätzlich noch alle zwei Tage läuft :)

opened a bug for this btw: https://bugzilla.proxmox.com/show_bug.cgi?id=6282

no, that shouldn't make a difference. upid files are created when a task spawns, the api will still spawn a prune task. you can really only try to shorten that file name. and yes `-` seems to get encoded, if you manage to remove those it might help.

ok well iirc correctly the name of the snapshot is encoded in the upid name, so a) there error message leaks the name unless you also altered it and b) it seems you are using something fairly long here, could you change that to something shorter? alternatively, you can use a shorter user or...

seems like the temporary upid file here is too long for this to work, you could try to run the prune command with the user directly instead of the token, that would shorten the file name. this is just a workaround though, we'll need to look into creating a proper solution here. ist there a...