Search results

Without the 'modprobe' word. The /etc/modules file is a list of modules to load at boot time.

Add them to /etc/modules.

Why the noout flag is set?

Try to add ipv6.disable=1 to the kernel boot command line (in /etc/default/grub).

Because anyone in the internet can send an UDP packet to your proxmox host and ports 5404, 5405, this will confuse the corosync process as you can see in the syslog message. You should protect your proxmox network with a firewall.

This is a very bad idea. And propably the cause of this syslog messages.

Try to use aptitude. It will fix a dependency problems for You.

You should buy a subscription or use a no-subscription repository. https://pve.proxmox.com/wiki/Package_Repositories#_proxmox_ve_no_subscription_repository

Are You going to use ceph luminous with bluestore? If yes, journaling device isn't necessary.

Public keys are public (known by everyone) by design. How can public information leak? Please show 'pvecm status' from node pve3.

With route blackholing only outgoing packets (from VM to blocked IP) are affected, so you will stay with many half-opened tcp session until they get tcp timeout.

Install pve5 from scratch, with different IP and hostname and add it to the cluster.

Full syslog attached. About hour before crash: Dec 16 07:48:28 p1 pmxcfs[2110]: [dcdb] notice: data verification successful

Hi. All nodes with bond interfaces, first eth in bond connected to master switch, second - to slave switch. When power back on slave switch: Dec 16 08:36:16 p1 kernel: [2926029.077939] bnx2 0000:01:00.1 eth1: NIC Copper Link is Up, 1000 Mbps full duplex Dec 16 08:36:16 p1 kernel...