[PVE-User] implementing openvswitch

Sten Aus sten.aus at eenet.ee
Wed Jan 28 14:09:03 CET 2015


I changed my interface names not to mix them up with linux bridges (as 
linux bridge was used by all of the VM-s, not just this particular 
proxmox host).

So, before i had (linux bridge) interfaces: vmbr10, vmbr15 and so on.
After, I had implemented ovs, I change them to OVSIntPorts called 
"backnet", "proxmox" and tagged to VLAN. But I just wanted to check - 
and now I've renamed all my interfaces with the same names they were 
used by Linux bridge and cluster is green! :)

Only thing - if anyone could speculate/help here:
If I manage to change all my nodes to ovs, can I rename interfaces then? 
Or is it more like "once you go vmbr10, you never go back"?


On 28.01.15 12:35, Sten Aus wrote:
> Hi
> My cluster is almost green (only this ovs node is red), but GUI is 
> working.
> Output from nodeB (linux bridge). (Changed and sanitized some values)
> cat /etc/pve/.members
> {
> "nodename": "nodeB",
> "version": 32,
> "cluster": { "name": "/greatcluster/", "version": 50, "nodes": /alot/, 
> "quorate": 1 },
> "nodelist": {
> /  /output omitted//
> *  "nodeB": { "id": 3, "online": 1, "ip": "***"},*
> /  /output omitted/
> /*  "nodewithOVS": { "id": 13, "online": 0},*
> /  /output omitted//
>   }
> }
> Output from nodewithOVS shows only this one node, no cluster information
> cat /etc/pve/.members
> {
> "nodename": "*nodewithOVS*",
> "version": 0
> }
> It's weird. I've only edited my interfaces file during this 
> OVS-process. Now I've restored my old Linux bridge configuration, but 
> still this node is red.
> Maybe my OVS is correct and working but some configuration issue? This 
> node has been switched off from cluster for about a week now.
> On 28.01.15 10:09, Sten Aus wrote:
>> Hi
>> Thanks for your reply! I read your topic, but all MTUs are 1500, only 
>> iscsi runs on MTU 9000.
>>> Do you see any other indications that anything is failing in the 
>>> cluster?  Things such as 'pvecm status' and 'clustat' - how do they 
>>> output?  When looking in the web interface are you able to see and 
>>> connect to all nodes?  Do they show a green or red status?
>> They output very slowly, but finally they show everything is OK. 
>> Cluster has quorum and all expected votes are completed.
>> One thing that GUI almost crashes after I connect this ovs interface 
>> to proxmox cluster. Everything is blank and login takes forever.
>>> I had one node configured with OVS and the other without and they 
>>> were able to talk for me without issue.  My problem came down to an 
>>> MTU mismatch between the nodes.
>> What MTU are you using and how is your topology? If you want, you can 
>> reply this in private!
>> All the best
>> Sten
>> _______________________________________________
>> pve-user mailing list
>> pve-user at pve.proxmox.com
>> http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
> _______________________________________________
> pve-user mailing list
> pve-user at pve.proxmox.com
> http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.proxmox.com/pipermail/pve-user/attachments/20150128/a9198eae/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3227 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://lists.proxmox.com/pipermail/pve-user/attachments/20150128/a9198eae/attachment.bin>

More information about the pve-user mailing list