Latest activity
-
Jjohnsonmelof replied to the thread PVE 9.1 with Kernel 6.17 - Unstable?.Hi, Are you still having these issues? I just updated my environment from 8.4 to 9.1, and since them, I'm having IO issues and random reboots. As mentioned before, the workaround looks to be pinning the kernel to version 6.14, but it isn't...
-
UdoB reacted to Johannes S's post in the thread Use Disk PassThrough or Raid-0 for PVE with Ceph (Testsystem) with
Like.
Regarding Ceph: The PVE wiki has a page how to set up a mesh network for three nodes. This saves some bucks because you will need less switches: https://pve.proxmox.com/wiki/Full_Mesh_Network_for_Ceph_Server The reference doc chapter on... -
UdoB reacted to Johannes S's post in the thread Is there a NON-host way to automate zfs dataset snapshot creation? with Like.
This makes it a prime candidate for a minimal alpine based container imho: With HA I always have a working version in the cluster
-
WFor anyone that pinned on the 6.14 the kernel: the 6.14.11-8-pve kernel (with Dirty Frag fix) seems to also be running without MegaRAID issues so far.
-
JJohannes S replied to the thread Is there a NON-host way to automate zfs dataset snapshot creation?.This makes it a prime candidate for a minimal alpine based container imho: With HA I always have a working version in the cluster
-
JJohannes S reacted to UdoB's post in the thread Is there a NON-host way to automate zfs dataset snapshot creation? with Like.
Never used it with this option in my mind. Although I thankfully recognized that I only need one instance of that script per cluster, so obviously it integrates into PVE :-)
-
JJohannes S replied to the thread Use Disk PassThrough or Raid-0 for PVE with Ceph (Testsystem).Regarding Ceph: The PVE wiki has a page how to set up a mesh network for three nodes. This saves some bucks because you will need less switches: https://pve.proxmox.com/wiki/Full_Mesh_Network_for_Ceph_Server The reference doc chapter on...
-
JJohannes S reacted to Martin.B.'s post in the thread Use Disk PassThrough or Raid-0 for PVE with Ceph (Testsystem) with Like.
Thanks for your input, Johannes S. This is exactly why we are testing as much possibilities as we can. And, at the moment, we are learning very much every day. In the past we would have installed a small VMWare cluster with shared storage...
-
Jcan you check the file-restore logs under /var/log/proxmox-backup/file-restore (on the Proxmox VE host)? this message appears usually when there is a filesystem in use in the guest that is not readable with a linux kernel also, please update...
-
MMarkusKo replied to the thread Ceph OSD crashes on I/O errors under light load with kernel 7.x - ssd via pcie adapter - FIXED with 6.17.13-7-pve.Try to update the firmware of your Samsung drives. Try to disable PCIe ASPM in BIOS or any other power saving feature. Check if the drives overheat -> improve cooling. Could also be an issue with the pcie -> m.2 adapter ...
-
Ssin3vil replied to the thread Cluster retransmit issues.Hi @kd-infradijon, Thanks for replying back. I can see from your thread that the solution was to force 10G (or 25G) links instead of auto-negotiation? We didn't have the ability to do this in our case, as the three existing nodes have 1G NICs...
-
Cchenwb replied to the thread [SOLVED] Are there any companies that extensively use PVE in daily production environments?.I Will install pve in 3 server
-
Cchenwb posted the thread [SOLVED] Are there any companies that extensively use PVE in daily production environments? in Proxmox VE: Installation and configuration.Are there any companies that extensively use PVE in daily production environments? Our data center currently utilizes commercial virtualization software. When issues arise, we rely on the vendor for support. However, the downside is the high...
-
Oothenone reacted to meyergru's post in the thread Moved SSD from one motherboard/pc to another -> Higher temperatures and fan speeds with Like.
Unless you use some Linux tools to set it explicitely, the actual power modes are determined by the BIOS. Both Intel CPUs should use around the same power when idle, the difference is the ACPI settings which determine how deep the sleep states...
-
gurubert reacted to torbho's post in the thread VMs freeze after node failure/reboot in 5-node Ceph cluster with Like.
After all we were unable to find a satisfactory solution with the ring topology. We kept experiencing the same issues over and over again. We have now switched to a star topology using a 10 Gbit switch. Since then, the problems have disappeared... -
gurubert reacted to tchaikov's post in the thread VMs freeze after node failure/reboot in 5-node Ceph cluster with Like.
Thanks for the detailed network diagram. Pulling together everything you've shared, I think the picture is fairly clear now -- though some targeted measurements would still help confirm the timing. The core reason VMs freeze During Ceph...
-
Oothenone replied to the thread Moved SSD from one motherboard/pc to another -> Higher temperatures and fan speeds.Alot of suggestions already, thank you! To answer some of the comments above: - I have not done any custom tuning on the previous server. And as @proxuser77 says, all suggestions are welcome, Proxmox specific ones, but also BIOS and hardware...
-
Oothenone reacted to proxuser77's post in the thread Moved SSD from one motherboard/pc to another -> Higher temperatures and fan speeds with Like.
So what? A few extra hints and tips won't do any harm, will they? ;)
-
Oothenone replied to the thread Moved SSD from one motherboard/pc to another -> Higher temperatures and fan speeds.Yes, this is exactly what i wanted clarified, thank you. And i can tell you, it's not clear to me with 1/2 year of experience with Linux, mostly in Ubuntu. Although, further help besides Proxmox related does not hurt, so all suggestions are...
-
Pproxuser77 reacted to Impact's post in the thread Moved SSD from one motherboard/pc to another -> Higher temperatures and fan speeds with Like.
Give something like this a look grep -r . /sys/class/powercap/intel-rapl:0/constraint* grep -r . /sys/devices/system/cpu/cpufreq/policy*/energy_performance_preference There might also be PL1/PL2 options in the UEFI of the device. My MiniPC...