Search results for query: memory usage

...migrations in general. From our perspective, it looks more like a VM-specific issue. The affected VM is heavily loaded in terms of memory usage, and during the migration process, the final memory switchover may fail because the operating system keeps continuously loading data into RAM...

...of System in VM setup to SeaBIOS selected: Didn't work. Fails at "installing kernel". Also when error report pane is shown, the memory usage shoots up to 90%. - Trying http://mirror.pilotfiber.com/ubuntu/ mirror at install console: Didn't work. Fails at "installing kernel". -...

...as used memory. Windows usually reports what you see in the Task Manager if configured properly but it sometimes zeroes all the memory (check with RAMMap) making it show as fully used. With PVE 9 you now also have information about Host Memory Usage. Many variables. Much potential for confusion.

https://pve.proxmox.com/wiki/Upgrade_from_8_to_9#VM_Memory_Consumption_Shown_is_Higher If this is a Windows machine, make sure that the Balloon Device is enabled (Hardware -> Memory, in the advanced part) and that the driver for that device as well as the "Balloon Service" is installed and...

Hi everyone, some time ago I updated my Proxmox host and noticed a strange behavior related to memory usage. Even when using VirtIO (balloon/IO), the host shows the memory as almost fully used, while inside the VM the memory usage does not exceed 30%. Has anyone experienced this issue or...

...in size, but in an inverse manner. Nodes with lower CPU usage have longer bars than Nodes with higher usage. "Nodes With the Highest Memory Usage": Same behavior as "Guests With the Highest CPU Usage" for all PVE hosts, but slightly larger for PBS (the usage is being listed as 47.85%, but the...

See https://pve.proxmox.com/wiki/Upgrade_from_8_to_9#VM_Memory_Consumption_Shown_is_Higher :

I recently noticed that Proxmox shows the memory usage of an OPNSense Virtual Machine at 100% consumed, in essence, out of memory. Yet, when I log into the Web UI of the OPNSense VM, it shows memory at only 23% consumed, and it is operating normally. What might be causing this discrepancy?

the package in question hasn't been bumped yet, once it is the referenced bug will be updated. yes, the overload might very well be on the storage level. the log indicates that both pvestatd and the ha loops/cycles take way too long.

it seems, but it is absolutely not, our normal cpu load during the day is kept very low, same with memory. This 'overloaded' is caused purely by io delay on the mounted backup volume, which makes sense is slower during backup windows. This should however not cause a complete PVE node to reboot...

20 mins before the last reboot and one prior to that.

...SWAP so the 512M given to the CT won't do anything. What you can do is to set up ZRAM on the node (5-15% or so) and give the CT 2G or so SWAP. Give the CT more memory. Perhaps 12-16G. If it still happens I'd contact the immich authors for help debugging the memory usage. Keep us updated though :)

THe memory usage graph would be of interest as well as per #7.

And the memory graph for the CT. I'm curious if the memory usage slowly rises (perhaps leaking) or if it's triggered by an event. Perhaps a nightly cronjob or similar that just needs more than 8G of memory to do its thing. Right now the memory usage looks okay. Note that this is mostly a immich...

I actually wanted to see the processes inside the CT too. Hence why I made the top command sort by memory usage. The node/CT Summary has graphs for memory usage and pressure statistics.

Yes, KSM is in use heavily. I noticed when debugging this via strace that it was walking the KSM memory pages or something, which seemed to be the majority of the 'wait' time. Once this fix is pushed through, do you think I will no longer need to do this RRD tuning?

...a fix that should improve the situation: qemu-server (9.1.4) trixie; urgency=medium [...] * fix #6935: vmstatus: use CGroup for host memory usage to avoid performance impact of reading the PSS metrics when there is a significant amount of kernel same-page memory being merged...

...28 core CPUs. We are in the process of migrating VMs onto the cluster from VMware (about 150 VMs so far) and are currently at around 30% memory usage on each hypervisor. CPU usage is low (less than 5%). At what point does it make sense to add more hypervisors? We usually keep one hypervisor...

...any intervention from our team. We did migrate it to Proxmox with the same configuration, and now we had to intervene a lot to help with memory usage. 1. The systemd-oom is triggered at one point and kill the application 2. The /proc/meminfo show that all memory is eaten up 3. To workaround...

...my NUC NUC8BEB - i5-8259U has been running much hotter, never dropping below 75°C and often reaching 95-100°C, even though with the 4 VMs and 1 CT running, it doesn't exceed 5% CPU usage or 33% memory usage. Do you think this could be also related to one of the recent kernel updates? Any ideas?