Search results

There are no 1 gig links anywhere. All other functions routinely exceed 500MBps.

Hi, I've recently upgraded my network stack to 10 gig. I've added "migration_unsecure: 1" to /etc/pve/datacenter.cfg and now live migrations routinely see over 600MBps. Writing directly from the hosts using dd to the nfs shared storage also sees 600MBps. However, when clicking on the "move...

Hi, I'm not looking for a hack. Consider this a feature request. I would like a GUI to be able to schedule tasks globally within a cluster. The technology already exists as backups are currently able to be scheduled globally within the GUI. Please expand on this and create a place in the GUI...

It's already done for scheduling backups. It doesn't seem like that much of a leap to do it to allow for other things.

Hi, It would be nice to have a global scheduler. Something at the Datacenter level that would allow me to schedule tasks for all hosts in a cluster. A "global crontab", if you will. Regards, micush

There is a script for this functionality at https://github.com/kvaps/pve-autosnap. It works pretty well. Schedule it as a cron job.

After testing, I answered my own inquiry. It is possible to do this. Use a bridged network device on the VM with no VLAN tag. Then inside the VM, linux in my case, create a network configuration with multiple VLAN tags on a single network interface. As long as the tag is allowed on the trunk...

Hi, Thanks for the response. However, I am wanting the VM to tag multiple VLANs from one NIC. Is it possible without assigning 10 NICs to a VM, each assigned to a different host bridge in a different network? Regards m

I would like to set up a .1q trunk to a VM, so that the VM does the VLAN tagging, not the host server. Is it possible?

Only the VMs assigned to a specific HA group will be able to use the hosts that belong to that HA group. It's probably easier to think of it as partitioning the HA cluster into different groups of hosts.

I have done exactly this. Create a new HA group. In that particular group only put the hosts that can manage the VM. Then add the VM to HA, selecting this newly created group. In the event of failure, HA will only move the VM to the hosts within the specific group.

There is no output on the 'netstat' command. A manual 'qm delsnapshot' gives a timeout error: ~# qm delsnapshot 108 daily201606031845 VM 108 qmp command 'delete-drive-snapshot' failed - got timeout

A little more investigation reveals some sort of timeout when running the script interactively: vm108: Removing snapshot daily201605311845 VM 108 qmp command 'delete-drive-snapshot' failed - unable to connect to VM 108 qmp socket - timeout after 5974 retries vm108: Removing snapshot...

Hi, The answers to your inquiries are as follows. - The script executes '/usr/sbin/qm delsnapshot $vm $snap' - A config file for one of the effected VMs is: #VM 101 agent: 1 boot: c bootdisk: virtio0 cores: 5 hotplug: disk,network,usb localtime: 1 memory: 16384 name: vm101 net0...

I have a script that automatically creates and deletes snapshots for all VMs in a FIFO fashion. For almost all of my VMs this works as expected. For some of my larger VMs I end up with something like this: I can then go in and delete the deleted snapshots by clicking on the 'Remove' button...

+1 to implement please

You can power off a VM on local storage and migrate it to another local storage host without issue. Why can't an online migration with local storage be done when an offline one can?

Sorry to resurrect an old thread. There is a script for this functionality at https://github.com/kvaps/pve-autosnap. It works pretty well. Schedule it as a cron job.

In Proxmox 4.2, if I attempt to add a firewall security group to a VM it does not allow me to select the security group. The group is there and can be chosen, but it is not allowed to populate the field. To reproduce, go to [Datacenter] -> [Firewall] -> [Security Group] and create a new one...

I had a similar issue. Turning off IGMP snooping on my switches solved the issue for me.