Search results

  1. D

    Ceph Dashboard (RADOS GW management problem)

    Hi, We have been using the Ceph MGR Dashboard to successfully manage S3 buckets and user accounts since Octopus (https://forum.proxmox.com/threads/pve-6-3-with-ha-ceph-iscsi.81991/). This continued to work with us migrating to Ceph Pacific and exchanging civetweb for beast, although we...
  2. D

    Loss of connectivity - OvS (apt-get -y dist-upgrade)

    PVE 7.3 results in loss of connectivity when OvS (Open vScwitch) is upgraded. OvS where 2 x 10G interfaces are bonded, vlan 1 is untagged for the node itself and vlan 100 is for Ceph and cluster communication: [root@kvm1a ~]# cat /etc/network/interfaces auto lo iface lo inet loopback auto...
  3. D

    Ceph Pacific (16.2.6) - Some OSDs fail to activate at boot

    This issue at heart is due to `blkid` forming an ambivalent opinion of the block device and udev subsequently not applying the ceph:ceph ownership of the device when it initialises. The root cause is as such that `blkid` finds other filesystem identifiers in the block device (as expected as the...
  4. D

    VM shutdown, KVM: entry failed, hardware error 0x80000021

    Herewith confirmation as well that running with TDP_MMU disabled resolves this problem for us. It appears to primarily affect Windows 2019 and Windows 2022 hosts. Also took us a while to identify as most VMs got hit by this after hours, when they are more idle than during office hours. The...
  5. D

    Debian update with open-vswitch stopped networking

    The following appears to work but is very hacky: Open two sessions, one to trigger a powerdown/up cycle in 5 minutes time and another where you run updates in a screen session: Enable SysRq, set a power on event in 5:20 minutes, sleep 5 minutes and instantly power off: echo 1 >...
  6. D

    Debian update with open-vswitch stopped networking

    This does appear to be an upstream Debian issue, what's odd is that Proxmox nodes don't fence... Sample Debian 11.2 VM where OvS also dies when running an apt-get update: [root@mininet ~]# cat /etc/network/interfaces # This file describes the network interfaces available on your system # and...
  7. D

    Debian update with open-vswitch stopped networking

    Connecting to the host out of band (iDrac) we can initiate a controlled restart, after which everything works perfectly. State after reboot: [admin@kvm1c ~]# pveversion -v proxmox-ve: 7.1-1 (running kernel: 5.13.19-6-pve) pve-manager: 7.1-11 (running version: 7.1-11/8d529482) pve-kernel-helper...
  8. D

    Debian update with open-vswitch stopped networking

    We were up to date as of Friday morning 7:11am (GMT+2), updates released since then result in OvS being broken when running: apt-get update; apt-get -y dist-upgrade; We're running ifupdown2 with /etc/network/interfaces as such: auto lo iface lo inet loopback auto vmbr0 allow-ovs vmbr0 iface...
  9. D

    Ceph Pacific 16.2.7

    Many thanks, that's excellent news! Continue to be pleasantly surprised by the rapid pace of development!
  10. D

    Ceph Pacific 16.2.7

    Hi, We are affected by a bug which has been merged and backported to Ceph Pacific, to be included in the next v16.2.7 release, which was released for testing as RC1 on the 3rd of December. There is additionally a data corruption bug which recommends that people not upgrade to Pacific until...
  11. D

    Issue with nested VM - KVM: entry failed, hardware error 0x7

    We had an almost identical issue with GNS3 VM (v2.2.27) where this VM had been updated to latest available packages in Ubuntu 20.04 (Focal Fossa). After updating the VM to Ubuntu 21.10 (Impish Indri) it's working as expected. To be clear, we're running the GNS3 VM where we updated Ubuntu to the...
  12. D

    Good way to modify corosync.conf in our cluster

    We mitigated against false positive fencing scenarios by making the following changes: Overview: https://forum.proxmox.com/threads/unexpected-restart-of-all-cluster-nodes.97021/#post-426562 Restart corosync if it ever crashes...
  13. D

    YubiCloud integration

    Edit: Apologies, I miss read the initial question and thought you wanted Yubico OTP. The forum link below however also details how to setup WebAuthn (U2F is deprecated and expected to stop working in Chrome next year March 2022). Navigate to https://upgrade.yubico.com/getapikey/ and apply for...
  14. D

    Ceph Pacific (16.2.6) - Some OSDs fail to activate at boot

    We have uncovered a problem with Ceph pacific OSDs not always starting automatically after a node is restarted. This is relatively prevalent with nodes exhibiting a single OSD with this problem approximately 70% of the time. We had one occurrence where a node had two OSDs in this state, whilst...
  15. D

    PVE 7.1 - U2F broken, confused about WebAuthn

    We upgraded our last PVE 7.0 cluster to 7.1 this weekend. Didn't need to download the packages from testing as there were already superceding ones available via the enterprise repo: [admin@kvm5b ~]# dpkg -l | grep -e libpve-access-control -e libpve-rs-perl ii libpve-access-control...
  16. D

    [SOLVED] CEPH IOPS dropped by more than 50% after upgrade from Nautilus 14.2.22 to Octopus 15.2.15

    We observed better overall performance and lower latency after upgrading to Pacific. We had another issue during this maintenance window where memory DIMMs were marked as failed and subsequently reduced available RAM. This led us down a garden path but Pacific is outperforming our experience on...
  17. D

    PVE 7.1 - U2F broken, confused about WebAuthn

    Hi Wolfgang, I installed the two packages and can confirm that I can now use WebAuthn to login to any cluster member server, as long as I remove the 'origin' line from the datacenter.cfg file: webauthn: id=company.co.za,rp=company.co.za I'm unfortunately not able to test legacy U2F until this...
  18. D

    PVE 7.1 - U2F broken, confused about WebAuthn

    U2F mult-factor authentication is unfortunately broken after upgrading to PVE 7.1: We had previously set the U2F AppID to point at a JSON document on a redundant web hosting service 'https://u2f.company.co.za/kvm1-appid', where this file contained a list of the possible facets that it would...
  19. D

    WebAuthn registration failed

    We've been using YubiKey 5 NFC for U2F but this breaks after upgrading to 7.1. We have a cluster of nodes with a hosted appid file, to make it work on any node: Cluster U2F configuration: [root@kvm1a ~]# grep u2f /etc/pve/datacenter.cfg u2f: appid=https://u2f.company.co.za/kvm1-appid Content...
  20. D

    PVE 7.0 BUG: kernel NULL pointer dereference, address: 00000000000000c0-PF:error_code(0x0000) - No web access no ssh

    Hi Thomas, I was under the assumption the numeric value before '-pve' in the package name 'pve-kernel-5.11.22-4-pve' was a revisioning number to differentiate packages produced by Proxmox. I subsequently hoped that Proxmox could increment this identifier to avoid situations where 'dpgk -l...

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!