they are no problem with PG autoscale, it's really depend of what you want to do with ceph. you could have multiple pools (in this case, need to define target ratio for each pool depending of what you want to do with your pool). you can have rbd...
here a small blog article about new fast erasure coding implementation && partial read , and some benchmarks inside too
https://ceph.io/en/news/blog/2025/tentacle-fastec-performance-updates/
The main problem with PG autoscale, by default, if your pool is empty, it'll not set a lot of PG, and this can slowdown write a lot (because of pg lock).
You should have between 100~200 PG by OSD (you can see them osd by osd on the osd tree)...
@alpha754293
- how many PG do you have on your ceph cluster ? (new pool on empty storage are setting low number of PG by default with target ratio is not configured on the pool, and this can give very low write speed).
- do you have enable...
Thanks daanw, much appreciated! I ran the following command and will keep monitoring PVE for any potential hardware unit hangs.
ethtool -K eno1 gso off gro off tso off
Hi,
I have wrote a script to generate an iso install with debian + proxmox package merged + debian preseed file for automatic installation,
with crypto lvm + specific partitions for CIS compliance.
It's using secureboot with proxmox shim &&...
If you don't like the nag screen and don't want to contribute to the development ( guess what developers need to pay bills too ) there is an easy fix: Don't use the software.
For cross-cluster migration you can also use the pct remote-migrate and...