Search results

Okay, basically it's exactly what's happening in the bugzilla report: I have a customer with a Proxmox Cluster who uses a shared Open-E iscsi Storage Cluster. These storage Clusters are somwhat strange and have some design issues, but until the latest patch there was a workaround for this...

So what does this tell me? I can't find a way to fix this. My next try would be to replace the perl modules by the ones of the previous commit, but this is not a update-proof solution, and i can't say more about other side effects.

is there any update on this? Verious iSCSI Cluster still won't work with Proxmox.

I have the exact same Effect on a 5 Node Epyc Milan / 7313P (Board Supermicro H12SSW-NTR) Cluster. Ran stable for a year, since 8.2/Kernel 6.8 random lockups of single Nodes after 1-3 days. Console freezes, no error Messages anywhere, just frozen. Rebooted all nodes today back to 6.5 - I will...

//update: This looks a whole lot better. Seems to sustain the LTO Drive speed except a very few dips. Thank you! regards.

We're getting somewhere: 2024-05-07T16:56:14+02:00: Starting tape backup job 'zfs:cephfs-mai:lto9:cephfs' 2024-05-07T16:56:14+02:00: update media online status 2024-05-07T16:56:16+02:00: media set uuid: c3d64c07-c811-48f0-9845-086cead14e55 2024-05-07T16:56:16+02:00: found 9 groups (out of 9...

.. build running. It just takes ages. I have v1 of your series + the fix. will that work?

Okay, so we can already rule out the Drive to not be able to sustain the specified 300MB/s.. so the bottleneck must either be the ZFS, or the Code that's pulling the data off.

I notice one thing, which is not totally logical: 2024-05-07T13:32:18+02:00: wrote 1272 chunks (4299.95 MB at 256.08 MB/s) 2024-05-07T13:32:37+02:00: wrote 1211 chunks (4305.72 MB at 255.80 MB/s) => 4305MB in 19s: 226,578947368 MB 2024-05-07T13:32:55+02:00: wrote 1490 chunks (4295.23 MB at...

What type of drive do you use?

Another observation: This is how it looks like in the Datastore graph - only the Tape backup job is running, but it constantly shows ~255MB/s: vs: 2024-05-06T20:01:25+02:00: backup snapshot "vm/11222/2024-05-05T20:35:05Z" 2024-05-06T20:02:05+02:00: wrote 7322 chunks (4295.75 MB at 183.45...

First I would like to provide you with some Specs: AMD EPYC 7313 16-Core Processor 12x Seagate Exos 20tb 4x 4tb NVME It's configured as raid-z3 with a 4-way mirror special device on nvme. It - seems - that the drop comes in after some time, a few minutes. I made another job...

More Update on the Topic: Increased the Thread to 16 (which is core count of the backup machine): 300MB/s (was <=200MB/s volatile) when running a tape job only. ~160-200MB/s (was <= 60MB/s volatile) when tape job was running while a verify job was running. //What I am noticing now is some...

Okay, was i impatient, built PBS myself and can conform significant improvements to the ZFS Performance on Spinners, especially when writing data to tape. Finally I can (at least if no other things are running) saturate the write performance of my LTO9 drive. Before the patch it maxed at...

Looking at the Thread again: If the gains are to visible for a single Spinner, the effect to have multiple IO threads should be even more Interessting on ZFS pools with several Disks...

Can I somehow test this without doing a complete PBS build myself? (eG. are there development deb's?)

This is awesome news, I am really looking forward to this..

So well, I investigated this a bit: PBS doesn't use rsyslogd by default, just systemd-journald. In theory there is the option to set LogFilterPatterns in the service definition for systemd for the proxmox-backup-proxy.server BUT: Bookworm has systemd 252, and the Option was introduced with...

Hi, i just noticed that PBS logging is very noisy: Mar 31 23:07:29 pbs-ba1-2 proxmox-backup-proxy[2878]: GET /chunk Mar 31 23:07:29 pbs-ba1-2 proxmox-backup-proxy[2878]: download chunk "/mnt/datastore/datastore/px11/.chunks/4583/4583ba2fb3e7c1086442e0> Mar 31 23:07:29 pbs-ba1-2...

hey @dcsapak, let me bump this up. Is there any news regarding tape write performance? thanks in advance.