Search results

I only found out about msecli from this ZFS benchmark thread and back then had not considered it for my benchmarks. So yes, I was wrong - it should be 4KB NVMe block size. And the default Ceph block size is 4MB - no idea if Proxmox does changes to the RBDs here.

The data for these graphs is collected by Zabbix agents into a Zabbix DB. From there I used the Zabbix plugin in Grafana. Our decision to use Zabbix is 10 years old and we moved away from Nagios. As long as we are still able to monitor everything (really everything!) in Zabbix we do not even...

I performance-tested from 1 to 4 OSDs per NVMe. It really depends on the system configuration - to drive more OSDs you need more CPU threads. See this thread and the posts around there. With my experience so far now I would just create one OSD per device. As Ceph uses a 4M "block size" I would...

So what magical and secret settings did you set then?

So here is the status... upgrade is in planning root@proxmox07:~# qm status 167 --verbose blockstat: scsi0: account_failed: 1 account_invalid: 1 failed_flush_operations: 0 failed_rd_operations: 0...

Ok, took some time to find out... proxmox-boot-tool does not prepare the systemd-boot configuration if /sys/firmware/efi does not exist - so to prepare the sda2/sdb2 filesystem for systemd-boot before booting using UEFI I had to remove those checks from /usr/sbin/proxmox-boot-tool.

So I was able to change the disk layout online by doing this: zpool status # !!! Be careful with device names and partition numbers!!!! zpool detach rpool sdb2 cfdisk /dev/sdb # Keep only partition 1 (BIOS), create partition 2 with EFI and partition 3 with ZFS fdisk -l /dev/sdb # Should look...

The VMs all reside on Ceph RBDs shared between three nodes. I need to change all three nodes. So from my point of view I think splitting the ZFS mirror, repartition, zfs send the contents and reboot should be the easiest.

Hi Dominik, so this is the output directly on the pbs01 This is from a local PVE cluster connected via 10G. On I have not yet experienced problems like mentioned above on it. And this is the output of a node of the 5-node Cluster. As mentioned the traffic goes via "The Internet". This...

Hi, I have a older installation of a 3-node Proxmox Ceph cluster - probably 5.3 or older. The OS disks look like this: So there is no 512M partition I could use for the EFI boot partition. I would like to switch to a UEFI boot, but how do I cut out that 512M UEFI partition?

Hi, we are trying to replace out current Backup solution with PBS. So I installed a PBS on an older ThomasKrenn 4HE Intel Dual-CPU RI2436 with some disks... One Proxmox Cluster is already running backups into this pbs01 for some time without problems. Now I try to run backups with our...

Welche zweite?

Might want to have a look here: https://forum.proxmox.com/threads/ha-nfs-service-for-kvm-vms-on-a-proxmox-cluster-with-ceph.80967/post-365321 But I can not recommend to run a nfs-kernel-server within a LXC container on Proxmox - as soon as the nfs-kernel-server hangs it might also hang your...

I can not recommend using nfs-kernel-server with a cephfs kernel client when using CephFS snaphots. As soon as the MDS service fails over from active to a standby MDS the NFS clients die with kernel panic or services running on them just die. It seems to be a caching problem. This problem...

nfs-ganesha runs in user space, so if it hangs or crashes in the CT (LXC) container it will not take the Proxmox Host down with it. If the nfs-kernel-server in a CT container has problems it might hang the Proxmox host. It is better to run a nfs-kernel-server within a Proxmox VM (KVM).

Deep scrubbing is not happening at all at the moment - and probably not untill this is finished... () 9 pgs not deep-scrubbed in time pg 4.76f not deep-scrubbed since 2021-01-31 13:12:33.368818 pg 4.72f not deep-scrubbed since 2021-01-31 16:48:10.470599 pg 4.649 not deep-scrubbed since...

One of the six remaining PGs finished today... and again two OSDs hat 100% Read IO usage as soon as the PG finished.

Yes, i can check them, but what should I be looking for??? root@proxmox07:~# ceph pg ls | head -n1 PG OBJECTS DEGRADED MISPLACED UNFOUND BYTES OMAP_BYTES* OMAP_KEYS* LOG STATE SINCE VERSION REPORTED UP ACTING SCRUB_STAMP...

No, the pg_num and pgp_num are now at 128 and I changed the autoscaler to "warn" for these pools. root@proxmox07:~# ceph osd pool ls detail pool 2 'cephfs_data' replicated size 3 min_size 2 crush_rule 0 object_hash rjenkins pg_num 128 pgp_num 128 autoscale_mode warn last_change 15199 lfor...

Adjusted settings after reading this. root@proxmox07:~# ceph config show-with-defaults osd.0 | grep 'osd_max_backfills\|osd_recovery_max_active\|osd_recovery_sleep' osd_max_backfills 3 mon osd_recovery_max_active...