Search results

You need to change the disk yes. No doubt about it

I'd rather say the problem is with the Current_Pending_Sector and Offline_Uncorrectable (both being > 0). In any case, /dev/sdb is physically damaged. Data recovery can be anywhere between hard and impossible

Hi. Still trying to get use of PVE hooks. I found that, during a live migration pre-start and post-start are fired on the destination node. But, when the migration ends, the pre-stop and post-stop are not fired on the source node. Is that expected or a bug ?

Yeah ! I found that I needed to close STDOUT and STDERR, but was missing the setsid part. Many thanks !!

This exact code will block the hook (just tried)

This is where I'm lost. My hook is a perl script, and running in a simple fork() still blocks the hookThis is where it's getting strange. In my hook, I have something like this : $SIG{CHLD} = 'IGNORE'; if ($phase eq 'post-start') { my $pid = fork(); if ($pid) { print "Will run...

For my need (see https://forum.proxmox.com/threads/running-hooks-when-incoming-migration-is-finished.54330 for some background), I need to run a hook script when a VM ends an incoming live migration. I'm trying to use the post-start hook. The idea : in the post-start hook, loop to see if the VM...

Indeed, I was only working from the cmdline and didn't even think about checking tasks logs. I'll use Sys::Syslog to have the message I'm intereted in in syslog. Thanks !

Thanks, I'll look into this and report here :-)

MMmmmhhh, in fact, stdout is captured, and can be seen in qmeventd logs (journalctl -fu qmeventd). But, only post-stop hook is fired. No trace of pre-start, post-start, pre-stop. I've simplified my script to be only #!/usr/bin/perl -w use strict; use warnings; print "GUEST HOOK: " . join(' '...

Hi. On some VM, I need to exec a hook script when an incoming live migration is over (and another one when an outgoing live migration is over). In my case, it's to redirect public IP from my hoster (Online.net) to the new node where the VM is running, and updating routes to setup ARP proxying...

Hi. I'm trying to use hook script for a VM. I've enabled the local storage to handle snippets, I put my script in /var/lib/vz/snippets/, with exec permissions, I've configured a test VM to use with qm set 150 -hookscript local:snippets/vm_hook.pl But it looks like it's not executed. The...

Or, to be fair, I strongly recommend monitoring of the allocation (both data and metadata) when using anything which does thin allocation

You can create a Linux bridge to do that (can be done from the GUI or by editing /etc/network/interfaces)). If you don't enslave a real NIC of the hypervisor to that bridge, it'll be isolated from the outside. And if you don't assign it an IP address on Proxmox, it'll be isolated from the host...

SMART is supported. But you're using an hardware RAID controller, so /dev/sda is not a single disk, it's the RAID array as seen by the OS. The array itself doesn't support SMART, but the independent disks does, and you can query them with something like smartctl -d megaraid,5 -A /dev/sda...

One of your disks is dying. You should get a replacement. If you want more details with smarctl, you need to use something like smartctl -d megaraid,5 -A /dev/sda Where 5 is the disk ID (seems to be disk 5, but you can check from 0 to 20 as sometimes they do not get contiguous ID. You can...

Well, I'd say the backend advertises tls support, but without implementing it correctly. Any chance you could just disable TLS on the backend ?

Most likely your internal mail server is misconfigured. Postfix will only try to use TLS if the remote end announces it support it. So it looks like your internal email server says : "Hey, I support TLS", but don't "work" when TLS is used (hard to tell more without logs)

Having 2 - 2 nodes would be even worst. Keeping appart the fact that this is cheating because you have only one real node (and when rebooting it, you loose 2 votes for example), if you have 2 - 2 nodes, no side would be able to get the qorum, and all the nodes would self fence

You'd need the qorum in the DC. In a HA setup, it's not only expected, but also a strong requirement to kill the nodes which do not have the qorum. Can't you move the heartbeat in the DC near the main server ?