Search results

Das lässt sich im Moment nicht besser realisieren. Die PGs werden ja dann auf Datacenter-Ebene verteilt, jede OSD ist hier gleich beansprucht, lediglich der "Flaschenhals" wäre der eine Server mit 6 statt einer Platte. Alle Server sind jedoch mit 10G angebunden und CPU-technisch gut...

Ceth ist 100% healthy, OSDs (SSDs) alle mit smartmontools überprüft, hier ein Screenshot von den OSDs

Moin! Wir betreiben einen großen CEPH / Proxmox Cluster und ich finde leider im Moment keinen Ansatz zur Problemlösung der Performanceprobleme. IOStat zeigt extrem hohe Auslastungen einzelner RBDs, wenn ich in einem Pool einen Ceph-Benchmark laufen lassen geht ein bestimmtes RBD an die Wand...

The problem is it exists only in newer kernel versions. We're running version 4.15.18-4-pve #1 SMP PVE 4.15.18-23 (Thu, 30 Aug 2018 13:04:08 +0200) because all kernels released afterwards are faulty, although we could not find any changes in your git repository concerning our observed bug...

Hello again, we really need some help now. We're paying a huge amount of money to Proxmox per year and would like to have this issue solved. We cannot upgrade _any_ machine and are stuck with an old version. Please help us. Appreciated.

We've tested the latest kernel right now, same problems: SCP transfer (and other kind of communications) from one LXC to another shows stalling transferspeeds after X amount of transferred bytes. We'd highly suggest to look into this as this affects all of our current cluster setups and we're...

We plan to upgrade our test-cluster to -7 on monday. We'll report our investigations regarding the mentioned problems afterwards.

For LXC its pct unlock <id> and the reason for the lock is simple, backup failed while vm has been locked and you have to manually unlock it to check whats up with it.

Hello everyone, we're encountering a massive performance issue once we've upgraded all our hosts to the latest kernel "4.15.18-5". The prior kernel was working fine (4.15.18-4). The problem persists and is reproducable once we're booting into the "...-5"-kernel. Symptoms: - VM becomes...

Appreciated. We've upgraded RAM and disabled Swap :)

Hello everyone, i'd like to ask for a best practice solution for our current setup. We're running a Proxmox-Cluster over 3 datacenters, on Hypervisors with SD-Cards (32GB each). Our problem is as follows: After testing LXC-Containers for our production enviroment, we've figured out, that even...

Hello everyone, i'd like to ask for help regarding a problem i recently got my hands on. We're running 3 Proxmox clusters over 3 datacenters. Backup routines run by night for all 3 clusters. Backups are done via CIFS and also NFS. From time to time i'm running into a problem, where the...

IMHO its a bug, because its defined to be multiline-compatible by design. Will open a bugreport though. /Edit: For reference purposes: https://bugzilla.proxmox.com/show_bug.cgi?id=1572

Dear Devs, This time i need some help :) In KVM/libvirt its common to be able to set the system/bios/product information. The problem we encounter is, that this information can be multilined (one line for each part of the information). Proxmox only allowes one string. For licensing purposes we...

The manual states for creating an OSD with filestore: If you want to use a dedicated SSD journal disk: pveceph createosd /dev/sd[X] -journal_dev /dev/sd[Y] But we're missing a certain flag here: -bluestore 0 So it should be: pveceph createosd /dev/sd[X] -bluestore 0 -journal_dev /dev/sd[Y...

In PVE-5.1 Bluestore is the default for CEPH. Wondering why we got an extra flag "Bluestore" in the GUI, shouldnt this be "Filestore" instead to use the legacy option? Just wondering, because this flag doesnt change anything at all.

Any today this: [ 735.167063] Uhhuh. NMI received for unknown reason 2d on CPU 0. [ 735.167255] Do you have a strange power saving mode enabled? [ 735.167437] Dazed and confused, but trying to continue [ 746.545439] Uhhuh. NMI received for unknown reason 3d on CPU 0. [ 746.545644] Do you...

Hey guys, we're running Proxmox (4) on multiple Lenovo Flex x240 m5-nodes with Dual E5 CPU and 10GbE. From time to time multiple nodes just go nuts. Maybe someone can lighten up my thoughts towards any solving idea Complete Bootlog: https://pastebin.com/raw/Wz2Ky8v8 Is this maybe related to...

Just to let everyone know: This bug affects _all_ Qemu versions after "detect_zeroes" has been implemented. This makes a guest able to DoS the host. To be safe, disable "detect_zeroes" in the meantime. I reported the bug to upstream. https://bugs.launchpad.net/qemu/+bug/1687653 Regards!

I think i've done more than enough to get this bug reported. I expect the devs to look at this forum as i dislike mailing lists.