Re: [PVE-User] Mellanox ConnectX-5 and SR-IOV

* Uwe Sauter [200514 22:23]: [...] > More details: > > I followed these two instructions: > > https://community.mellanox.com/s/article/howto-configure-sr-iov-for-connectx-4-connectx-5-with-kvm--ethernet-x > > https://community.mellanox.com/s/article/howto-configure-sr-iov-for-connect-ib-connect

[PVE-User] Mellanox ConnectX-5 and SR-IOV

Hi all, I had to change the hardware of one of my Proxmox installations and now have the problem that I cannot configure a Mellanox ConnectX-5 card for SR-IOV/passthrough. To be more precise I can boot the VM and it also recognizes the Infiniband device but I'm unable to assign a Node GUID and

Re: [PVE-User] Some VM's are not able to start

--- Begin Message --- The info is in the first line of the log "kvm: -device vfio-pci,host=:5e:00.0,id=hostpci0,bus=pci.0,addr=0x10: vfio :5e:00.0: failed to open /dev/vfio/46: Device or resource busy" This means the device is already passed through to another running VM, or it is being l

Re: [PVE-User] Some VM's are not able to start

Hello, Thank you for your support. But I am getting below error message after removing the 'hugepages =2' line in the VM config file. kvm: -device vfio-pci,host=:5e:00.0,id=hostpci0,bus=pci.0,addr=0x10: vfio :5e:00.0: failed to open /dev/vfio/46: Device or resource busy TASK ERROR: start

Re: [PVE-User] unable to hotplug cpulimit; cpu.* not in /sys?

--- Begin Message --- Apologies for replying to myself, but I have a litte more information. ‐‐‐ Original Message ‐‐‐ On Tuesday, May 12, 2020 1:12 PM, wrote: > Hi PVE-users, > > Lately, on a no-subscription Proxmox (that I update regularly since release > 6.0), I get the following erro

Re: [PVE-User] Some VM's are not able to start

--- Begin Message --- Remove the hugepages line from your vmid.conf (ie 100.conf) On Thu, 14 May 2020, 17:24 Sivakumar SARAVANAN, < sivakumar.saravanan.jv@valeo-siemens.com> wrote: > Thank you so much. > > What is the steps to disable the hugepage ? > > > Best regards > SK > > On Thu, May 14,

Re: [PVE-User] Some VM's are not able to start

The preferred method to disable Transparent HugePages is to add "transparent_hugepage=never" to the kernel boot line in the "/etc/grub. conf" file. The server must be rebooted for this to take effect. --- Gilberto Nunes Ferreira Em qui., 14 de mai. de 2020 às 13:24, Sivakumar SARAVANAN < siva

Re: [PVE-User] Some VM's are not able to start

Thank you so much. What is the steps to disable the hugepage ? Best regards SK On Thu, May 14, 2020 at 6:20 PM Mark Adams via pve-user < pve-user@pve.proxmox.com> wrote: > > > > -- Forwarded message -- > From: Mark Adams > To: PVE User List > Cc: > Bcc: > Date: Thu, 14 May 20

Re: [PVE-User] Some VM's are not able to start

--- Begin Message --- Do you really need hugepages? if not disable it. On Thu, 14 May 2020 at 17:17, Sivakumar SARAVANAN < sivakumar.saravanan.jv@valeo-siemens.com> wrote: > Hello Daniel, > > Thanks for coming back. > > I mean, I am unable to power ON the VM until shutdown the other VM's in t

Re: [PVE-User] Some VM's are not able to start

Hello Daniel, Thanks for coming back. I mean, I am unable to power ON the VM until shutdown the other VM's in the same host. There are 6 VM's running on each Host, But sometimes all 6 VM's would run without any issue. But Sometimes if I stop ( Shutdown) and Power ON ( Start) getting an error say

Re: [PVE-User] Some VM's are not able to start

- Le 14 Mai 20, à 17:38, Sivakumar SARAVANAN sivakumar.saravanan.jv@valeo-siemens.com a écrit : > Hello, > > We have implemented the Proxmox VE in our environment. > > So each server will have a maximum 6 VM. But not able to start the few VM's > ON until we bring down the 1 or 2 VM's

[PVE-User] Some VM's are not able to start

Hello, We have implemented the Proxmox VE in our environment. So each server will have a maximum 6 VM. But not able to start the few VM's ON until we bring down the 1 or 2 VM's in the same Host. What could be the reason. Kindly suggest. Thanks in advance. Best Regards SK -- *This e-mail mes

Re: [PVE-User] critical HA problem on a PVE6 cluster

Hi Hervé, Glad to read this :) Cheers El 14/5/20 a las 16:48, Herve Ballans escribió: Hi Eneko, Thanks again for trying to help me. Now, the problem is solved!  We upgraded our entire cluster in PVE 6.2 and now all is optimal, including HA status. We just upgraded each nodes, didn't change

Re: [PVE-User] critical HA problem on a PVE6 cluster

Hi Mark, Thanks. Yes we are investigating with network engineers. We upgraded the entire cluster in PVE 6.2 and the cluster is fully operational now. But we think indeed that something in the network has changed and caused the problem (switch upgrades ?) Therefore, for example, does activa

Re: [PVE-User] critical HA problem on a PVE6 cluster

Hi Eneko, Thanks again for trying to help me. Now, the problem is solved!  We upgraded our entire cluster in PVE 6.2 and now all is optimal, including HA status. We just upgraded each nodes, didn't change anything else (I mean in term of configuration file). Here, I'm just stating a fact, I