Search results

While there are some staff members actively supporting this community-forum (thanks for that!) the recommended way to be more visible to the developers is to put it into the official bug tracker: https://bugzilla.proxmox.com/enter_bug.cgi?product=pve Please cross-link from there to here and...

Setze "swappiness" auf Null. https://pve.proxmox.com/wiki/ZFS_on_Linux#zfs_swap - auf derselben Seite wie bereits oben verlinkt. Allerdings würde ich das nicht machen, sondern eine kleine Menge präsent lassen. Das basiert lediglich auf meiner persönlichen Wahrnehmung. Ärgerlicherweise kann...

Der "übliche Verdächtige" ist natürlich der ZFS ARC. Der belegte "früher" per Default bis zu 50% des RAM. Und da der ARC zwar schrumpfen kann, dies aber nur sehr langsam funktioniert, kam es dann gelegentlich zu Engpässen. Der aktuelle Default ist wesentlich kleiner - und man kann das natürlich...

In doubt of the details look into the good ol' man page: man zpool-attach SYNOPSIS zpool attach [-fsw] [-o property=value] pool device new_device ... If the existing device is a mirror or plain device (e.g. specified as "sda" or "mirror-7"), the new device will be mirrored...

That's "old behavior", running up to version 8.4 Upgrade to a current version (read: 9.x) and the information shown should get updated on every boot.

It is especially the technical readability of the data chunks from the physical disk which is confirmed by this process. (And then to be sure that the read data is actually correct the checksum is used.)

First: if it fulfills your requirements then it is absolutely fine. But: you are exchanging performance for other features. For me these are important and you have dropped them: technically cheap snapshots, transparent compression, guaranteed integrity - including self-healing and bitrot...

Is that a question ...? Sure. Simplified: for reading data it should be nearly the same as data is read from all devices in parallel, but for writing data you now have only a single vdev instead of two. The effect is that IOPS (and bandwidth) may go down 50%.

The datastore configuration on the PBS has a tab "Options". There you'll find "Verify New Snapshots". Probably that's the relevant toggle.

Verifying means to actually read all relevant chunks, calculate its checksum and compare that with another copy of the checksum which was stored persistently when the backup was created. It is often stated that PBS needs some IOPS. My suspicion: you see the effect of a slow storage device...

To give a single VM 83 % of the system RAM may be problematic, sometimes. I would start with 30 GiB and then 40 GiB - and I wouldn't be surprised if that works well. Sidenote, probably not relevant for that "Oops": that CPU has eight "Performance Cores". Start with these, not with 26.

You did not tell us much about your hardware. But this line: calls for: have you run memtest86+? If not: run it over night, at least several hours. At least two or three full "cycles". How much RAM does the system have? Can you start a Linux machine with that amount of Ram? (And use that RAM...

That's... not good, it is worst case! I would give it max. 14 cores and call it a day. Of course: ymmv. You have the hardware available --> just test it ;-)

I would be satisfied if there would be a "pve-cluster-sanity-check"-script which I could run from time to time, be it manually or via cron. It would work like the well done pve8to9 and check for some correct configuration settings. For example "relevant networks present on all nodes", "all...

Im Sinne "pro ZFS"? Klar doch: transparente Kompression, technisch billige Snapshots, Flexibilität beim Erweitern, nicht an die spezifische Controller-Hardware gebunden, garantierte Integrität und sicher weitere mehr...

Kann sein. Muss nicht. Du kannst diese Vergrößerung problemlos in mehreren Schritten vornehmen. Du kannst aber nicht einen Schritt zurück, also verkleinern. (Naja, auch das ist möglich, aber mit einigem Aufwand verbunden, der schlicht wegen der innewohnenden Komplexität "gefährlich" ist.)...

You see..., there are completely different (but valid) ideas, depending on who you ask - and that's a good thing! :-) My own version is to use the good old dd. Or most of the time its more featured bigger brother ddrescue. For operations like yours I like to keep it simple. Please follow...

This one was updated several times during the last 12 months: https://github.com/Corsinvest/cv4pve-autosnap For me it works as documented and these snapshots are fully integrated in the PVE-Webgui too. --- For non-PVE ZFS datasets I use "Syncoid" via https://github.com/jimsalterjrs/sanoid

I don't think so: https://blog.westerndigital.com/wp-content/uploads/2020/10/wd-red-family.png But yes, it should get replaced, in my opinion. With enough redundancy on the vdev level you could run a "long self-test" and then wait for "Currently pending" rising again. But that's a dangerous...

Ist dies ein Cluster? Falls ja: pvecm status bitte. Ist "/etc/pve" gemounted? (Das ist eine per fuse gemountete Datenbank!) Nur als Vergleich, bei mit sieht die Datei so aus: /etc/pve# ls -al user.cfg -rw-r----- 1 root www-data 3338 Dec 18 12:25 user.cfg Laufen alle relevanten Dienste? Bei mir...