Recent content by Felix.

Hetzner Datacenter in Falkenstein. Back then the pvestatd service was quite latency sensitive, which caused the PVE GUI to break if the storage did not respond quickly.

Works smooth like butter. Anyway, back when this thread started the cloud pbs was barely usable from germany, now it's just a (push) sync job running in the background, not affecting PVE in any way. Very happy to finally use this free offsite backup offer again!

Nutzt du den Server gleichzeitig als PBS (Backup Server) ? Falls nein, spar dir die Special Devices, das bringt allenfalls noch irgendwas, wenn du viele LXC container nutzt, da die ebenfalls ZFS als Filesystem nutzen dann. VMs mit Block Devices (ZVOL) wird das nicht nützen, die wären mMn. mit...

I'd go the route with the Remote PBS on VPS until S3 got stable. What kind of issues did you have with that setup?

This! Secure Erase is nice and fancy, but only works on somewhat still working disks. o_O Thats why Distributors offer services like KYD (Keep Your Drive) to customers caring about their data (or being legallcy forced to care :rolleyes:) Used LUKS + ZFS years ago, with a thumbdrive connected...

USB drives will cause pain for sure, one day or another. And screw SATA DOMs, Proxmox is logging quite a lot. You say you have a proper disk bay available, use that one. Sounds like you are running a Cluster anyway, so I'd recommend to choose a proper Datacenter SSD as Boot Drive and not try...

This assumes a reliable sending mail server that honors certain standards. Feeling better with having the buffer on my side, and I don't have to wait several hours after my own Mail Server is working again, I can directly deliver queued Mails from the PMG UI. I'd disable as much filtering in...

Interested in this as well. Want to ditch Plesk as Mail Server and replace it with Mailcow. When hosting at home, things may break, especially when we play around. PMG is hosted externally and acts as a Queue for incoming mails whenever my Homelab is unavailable, for whatever reasons...

Ahh, alright, that will be fine for a smooth and fast backup. I initially thought you are backing up over WAN. With some 100gb daily delta the remote syncs should be quick as well. The thing with NVMe is that you need the PCIe lanes, older platforms simply don't provide that many. Best you can...

So, you have no onsite backup and it backups live over WAN? Or do you have dark fibers there? What is the bandwidth and latency between the datacenters? I'm surprised this works at all without backup jobs failing or machines freezing frequently. Did direct backups over WAN before, but not at...

Did something like this some time ago, with 8 10TB drives, but using ZFS and dRAID RAID6 and a special device mirror (ab)using the rest of my boot drives, I partitioned them to only use 32gb for the OS. With a little tuning I got very acceptable results in GC and Verify Jobs, but that machine...

Yep. Most likely a BIOS issue - ram timings / speeds, maybe OC profiles misconfigured. Or simply a hardware issue, depending on the beep codes. Google those beep codes for your board, that should give you a clear hint what is wrong. When your system gets bootable again, like beside being able...

Years ago I had to deal with an MSA2050, in my old thread some multipath configs were posted, maybe they help you getting started. Also used SAS controllers. https://forum.proxmox.com/threads/pve-6-0-hpe-msa-2050-sas.57317/ (german though, but the config is probably most relevant part anyway)

Did you try a reboot? I don't remember exactly if there is a cache or something like that, there was some talk about this a while ago. Can you also share some screenshots of your PBS Webinterface, of the datastore view for example?

My point was about the directory not being filled with "a lot" of data, yes, the ton of directories, but not necessarily chunks. I considered the chunkstore containing only the directory tree as "empty", pardon my vague wording there. After he copied the data over, I somewhat expected the size...