Recent content by manu

1 0 6613 2 20 0 0 0 io_sch D ? 0:02 \_ [txg_sync] This tgx_sync here means ZFS is not able to write flush your IO to the disks fast enough for what you ask him to write. So the system is probably just waiting for the disks to respond. I am surprise to see this...

Usually systems "hangs" boil down to slow IO. You can try the command the command ps faxl and look if you have processes in the "D" state ( uninterruptible sleep ). Also have a look if you have in your kernel messages if you have the message "task blocked for more that 120s" and note...

What you probably want is called " PCI Passthrough" in the KVM world. see https://pve.proxmox.com/wiki/Pci_passthrough

yes this is possible, see https://pve4.intern.lab:8006/pve-docs/chapter-pvecm.html#_remove_a_cluster_node you have on to execute the delnode command on the node you want to keep, giving as paramater the nodes you want to delete

For connecting a SAN to Proxmox. * create a large LUN on the SAN side * add an ISCSI LUN using Datacenter -> Add storage * add a Volume Group on the LUN using Add VG -> Use existing Base Volume * this way you can use a on the large LUN LVs for VMs, without having to create LUNs for each VMs...

OpenBSD VMs are known to hang from time to time on Qemu/KVM. This was discussed on the openbsd mailing lists here: http://openbsd-archive.7691.n7.nabble.com/Openbsd-6-1-and-Current-Console-Freezes-and-lockup-Proxmox-PVE5-0-td322999.html People tried to mitigate the issue with a serial terminal (...

@lankaster Tried your setup here and backup works fine. This is my relevant vm.conf scsi0: pvepool_vm:vm-610-disk-2,cache=writeback,discard=on,size=9G Also using virtio-scsi, ceph luminous and PVE 5.1

Most Probably the host kernel killed the samba process because it was using more memory that the whole container 103 was allowed to use: * [ 1044.690344] Memory cgroup out of memory: Kill process 17172 (smbd) score 23 or sacrifice child * Task in /lxc/103/ns/system.slice killed as a result of...

the critical factor for ceph is that you need 10GB networks between the nodes, preferably on a dedicated NIC that's one of the most important thing to consider.

First do you see the link led "on" on this card ? This should work wheter the driver for this card is loaded or not. If yes, try to see if the kernel did associated a driver with the device with the command: lspci -k | sed -n '/Ethernet/,/Kernel modules/p' This is an example output on my...

If you use DHCP, then you can configure the DHCP server to send a fixed IP address corresponding to the Mac Adress of your container. You can get the IP address of a VM (not LXC ) via qemu-guest-agent, but this only works with Linux VMs (if you build the agent yourselt it will work for windows...

For LXC containers, the IP adress you set in the configuration are then applied inside the LXC container. So it is them same in the end ( NB: in your shell script you also read the IP adress from the configuration) As long as your containers are not using DHCP, this should work. If not, please...

Hi Vinny For such a task I would use one the API client library. You can either use proxmoxer ( https://github.com/swayf/proxmoxer) if python is your thing or our own libpve-apiclient-perl, that you can install on any debian based system ( no need to be a PVE system ) When you use these...

that can be really a lot of VMs for such a small pipe. Also reading the LACP entry in wikipedia: "This selects the same NIC slave for each destination MAC address" which means all your outgoing writes to the iscsi target might go through 1 single GB link .... I think you system is in danger...

the VMs remounting their volumes as read only is an indication that you're trying to push to much IO through the pipe I would advise you to monitor the latency while doing a restore ( with ping) and the io latency with ioping how much VMs do you have and which link to do you have to your...