Issue with Proxmox VE 3.0 when adding LVM group

[thias_]

Runnig Proxmox VE 3.0, I've added 4 LUNs from 2 ZFS Appliances connected using Fiber Adapters.
Two VG with 2 LUNs each have been created. (see below the outputs of pvscan and vgscan)

When trying to add the LVM group using the web interface, no Volume Group are discovered. The "Scanning" ends with no return.
Note that when only two LUNs (one VG) were connected to the Proxmox node1, the VG has been displayed and added using the web management interface.

In the syslog of the node, the following error is logged:

Jun  6 10:42:32 node1 pveproxy[185175]: WARNING: proxy detected vanished client connection
Jun  6 10:42:33 node1 pveproxy[242789]: problem with client 172.16.8.14; Connection timed out
Jun  6 10:42:38 node1 pveproxy[185175]: problem with client 172.16.8.14; Connection timed out
Jun  6 10:42:40 node1 pveproxy[242789]: WARNING: proxy detected vanished client connection
Jun  6 10:42:43 node1 pveproxy[185173]: problem with client 172.16.8.14; Connection timed out

Can you please help me on this issue.

Thanks and regards
--
Mathias


Output of pvscan:

  PV /dev/mapper/3600144f0fb390868000051af5887001b   VG testzone2-vg   lvm2 [10.00 GiB / 10.00 GiB free]
  PV /dev/mapper/3600144f0e7412ae1000051af586a001b   VG testzone2-vg   lvm2 [10.00 GiB / 10.00 GiB free]
  PV /dev/sda2                                       VG pve            lvm2 [135.62 GiB / 16.00 GiB free]
  PV /dev/mapper/3600144f0fb39086800005194aa7d0012   VG testzone1-vg   lvm2 [10.00 GiB / 200.00 MiB free]
  PV /dev/mapper/3600144f0e7412ae100005194aa0a0011   VG testzone1-vg   lvm2 [10.00 GiB / 10.00 GiB free]
  Total: 5 [175.61 GiB] / in use: 5 [175.61 GiB] / in no VG: 0 [0   ]

Output of lvscan:

  ACTIVE            '/dev/pve/swap' [17.00 GiB] inherit
  ACTIVE            '/dev/pve/root' [34.00 GiB] inherit
  ACTIVE            '/dev/pve/data' [68.62 GiB] inherit
  ACTIVE            '/dev/testzone1-vg/vm-100-disk-1' [9.80 GiB] inherit

 

[dietmar]

Maybe the storage is slow (timeout)?

 

[thias_]

The timeout occurs on the pveproxy side. I have the same server connected to the same storage but running Proxmox VE 2.3 and the error does not occur (Proxmox is able to list the VGs).

The same error message appears in the Proxmox VE 3.0 server syslog when tying to access https://IP:8006//api2/json/nodes/node1/scan/lvm

Is there any way to increase the timeout / keep-alive on the pveproxy side?

Thanks and regards

Maybe the storage is slow (timeout)?

 

[dietmar]

Ok, I filed a bug for that:

https://bugzilla.proxmox.com/show_bug.cgi?id=401

I will try to fix that for with the next upload.

 

[thias_]

Hi Dietmar,

Thanks fo taking care of this point.
In the meantime, I've been able to tune our storage connexion and get rid of the timeout.

Thanks again for your help and for the nice job done with Proxmox VE!
--
Kind regards
Mathias