spirit's latest activity
-
spirit replied to the thread corosync show link flapping (down/up) about every 3-4 minutes, but switch shows no problem.you can use "corosync-cmapctl -m stats" on each node to have detailed stats about latencies && message retry/loss -
spirit replied to the thread Cluster setup and HA configuration suggestions.you need to create 2 custom crush rules using the device_classes ceph osd crush rule create-replicated <crush_rules_name> default host <device_class> ex: ceph osd crush rule create-replicated myhddreplication default host hdd ceph osd crush...
-
spirit replied to the thread corosync show link flapping (down/up) about every 3-4 minutes, but switch shows no problem.do you have a granular monitoring (prometheus,...) which can check bandwidth every second ? don't trust average. do you have also checked your switch port buffer stats ? if it was a kernel bug, you'll have too on your 10gbit nic. I'm running...
-
spirit replied to the thread Create OVS Bond that supports MLAG.no, the mlag is only on your switch side. on pve host, it's a classic bond. do you have tried with linux bond instead ovs ?
-
spirit replied to the thread corosync show link flapping (down/up) about every 3-4 minutes, but switch shows no problem.but you have packet loss, this is even worse. (you can look at corosync stats too yes, definitively the recommandations is to have dedicated links for corosync https://pve.proxmox.com/pve-docs/chapter-pvecm.html but note that last...
-
-
spirit replied to the thread VM CPU issues: watchdog: BUG: soft lockup - CPU#7 stuck for 22s!.This could be related to memory fragmentation. do you use zfs ?
-
spirit replied to the thread RDSEED cpu flag deactivated after updating to kernel 6.8.12-20-pve.I don't think that microcode has never been in pe-firmware package. It's always the debian package. (and debian can be a little bit slow to update them)
-
spirit replied to the thread RDSEED cpu flag deactivated after updating to kernel 6.8.12-20-pve.you can try the firmware from debian testing: http://ftp.de.debian.org/debian/pool/non-free-firmware/a/amd64-microcode/amd64-microcode_3.20251202.1_amd64.deb (or upgrade your bios if your vendor provide it)
-
spirit replied to the thread corosync show link flapping (down/up) about every 3-4 minutes, but switch shows no problem.basically, if you have network saturation, the latency increase, and corosync remove the node from the cluster. (down/up don't mean that the physical link is flapping, it's simply that the node where you are checking the log, don't have a...
-
spirit replied to the thread yawgpp (Yet Another Windows Guest Performance Post).Well, you cant just pass all flags blindly. Some Windows versions can crash, some flags like tlbflush can have performance impact dépending of the workload,....
-
spirit replied to the thread corosync show link flapping (down/up) about every 3-4 minutes, but switch shows no problem.Is the 1gb link dedicated to corosync ? Link saturation could give flap like this.
-
spirit replied to the thread yawgpp (Yet Another Windows Guest Performance Post).The hv* flags are already send to command line by proxmox when you choose Windows as ostype.
-
spirit reacted to fiona's post in the thread [SOLVED] Applying pve-qemu-kvm 10.2.1-1 may cause extremely high “I/O Delay” and extremely high “I/O pressure stalls”. (Patches in the test repository with
Like.
From the analysis until now, the IO pressure seems to be a cosmetic issue or rather accounting issue in the kernel. QEMU switched to using io_uring for event loops with QEMU 10.2. The issue appears in combination with IO threads, where a blocking... -
spirit replied to the thread Proxmox cluster listening on all interfaces.Osd need to talk to monitor and clients (vm) which are running in public network, so they always use both public and private network
-
spirit replied to the thread Ceph - VM with high IO wait.please don't double post, https://forum.proxmox.com/threads/applying-pve-qemu-kvm-10-2-1-1-may-cause-extremely-high-%E2%80%9Ci-o-delay%E2%80%9D-and-extremely-high-%E2%80%9Ci-o-pressure-stalls%E2%80%9D-patches-in-the-test-repository.182186/page-2...
-
spirit replied to the thread Ceph - VM with high IO wait.the thing is that the host itself don't mount the rbd. It's done directly by each qemu process. the librbd is inside qemu, and this is the qemu process which is doing to connections to monitor && osd. (with kbrd it's different, as it's done by...
-
spirit replied to the thread EVPN SDN Multi-tenancy: Something similar to Inter-AS Option A?.if you have a vyos router, the best way is to peer with evpn with other proxmox nodes. (1 peer by vrf. 1 proxmox each zone is a vrf). if your coke tenant is a layer2 behind vyos, vyos should announce their ip/mac inside the evpn zone.
-
spirit replied to the thread Ceph - VM with high IO wait.It's really done when the qemu process is starting when librbd is initialized. (like other tuning that you can do in ceph.conf). It could be possible to pass params to qemu commandline, but qemu rbd driver have limitations of parameters that you...
-
spirit replied to the thread [SOLVED] VXLAN ARP timeouts.I 'm not sure that hook can work with live migration, as the arp entry need to be set after source vm stop (the arp entry need to be flushed from source horst) and before the resume of the target vm. This is a thing that we should implemented...
-
I finally solved problem thank u for help The root was my network devices configurated as stack and corosync was connected to different units, thats caused link flapping, when i connected to 1 unit problem was gone
