can't create XEN VPS with LVM volume

Submitted by SoftDux on Sat, 01/16/2010 - 05:18

I'm trying to setup a XEN VPS, but keep getting this error:

Copying 375.66 MB image file to host system usaxen05 ................................................................................................................................................................................................................................................................................................................................................................................................................. .. done, and added to host cache

Creating virtual system with Xen .. .. creation failed : LVM logical volume creation failed : Invalid RPC login to usaxen05

I can, however, create a XEN VPS with a file container, it's just the LVM ones that give me this error. What could cause this?

Status: 
Active

Comments

Submitted by JamieCameron on Sat, 01/16/2010 - 12:51

This could happen if the Cloudmin master system is somehow unable to contact Webmin on the Xen host system. Make sure that ports 10000 - 10010 are open on the host's firewall, and that if you select the host system from the left menu its status is shown as "Webmin".

Submitted by SoftDux on Sat, 01/16/2010 - 13:10

I can create a normal file based XEN VPS on this exact same host, and webmin does show on the left menu.

Submitted by SoftDux on Sat, 01/16/2010 - 13:11

I also disabled the firewall, and still get the same error:

root@usaxen05:[~]$ /etc/init.d/iptables stop Flushing firewall rules: [ OK ] Setting chains to policy ACCEPT: filter [ OK ] Unloading iptables modules: [ OK ] root@usaxen05:[~]$ iptables -L Chain INPUT (policy ACCEPT) target prot opt source destination

Chain FORWARD (policy ACCEPT) target prot opt source destination

Chain OUTPUT (policy ACCEPT) target prot opt source destination

Submitted by SoftDux on Sat, 01/16/2010 - 14:05

This only happens on this server, usaxen05 - on the other server usaxen03 it's working fine.

Submitted by JamieCameron on Sat, 01/16/2010 - 14:20

For usaxen05 , is the Webmin login set to something other than root? This can be viewed at Edit System -> Authentication options -> Webmin login mode.

Submitted by SoftDux on Sat, 01/16/2010 - 14:37

well, like I said, I can create a normal file based VM, so the authentication work fine. I can also see the server's resources, running VM's, etc. It's running Webmin 1.500, Cloudmin 3.8. And it's using "Login as root with password (the one I specified)" under Edit System -> Authentication options -> Webmin login mode > "Login as ... with password", and webmin port 10000.

When I click on "Open Wenmin on the left menu, it opens Webmin in a new window and I can see the resource usage and webmin menu

Submitted by JamieCameron on Sat, 01/16/2010 - 15:42

Only when an LVM-based VM is created is the Webmin API called on the host system, which explains why creation of a non-LVM instance works OK.

If you check the log file /var/log/secure or /var/log/authlog on the host system during the creation process, is any error message related to Webmin displayed?

Submitted by SoftDux on Sat, 01/16/2010 - 16:04

I don't see anything odd in /var/log/secure:

Jan 17 00:01:18 usaxen05 sshd[13128]: pam_unix(sshd:session): session opened for user root by (uid=0) Jan 17 00:01:18 usaxen05 sshd[13128]: pam_unix(sshd:session): session closed for user root Jan 17 00:01:21 usaxen05 sshd[13086]: pam_unix(sshd:session): session closed for user root Jan 17 00:01:22 usaxen05 sshd[13136]: Accepted password for root from 66.96.233.192 port 34453 ssh2 Jan 17 00:01:22 usaxen05 sshd[13136]: pam_unix(sshd:session): session opened for user root by (uid=0) Jan 17 00:01:23 usaxen05 sshd[13136]: pam_unix(sshd:session): session closed for user root Jan 17 00:01:24 usaxen05 sshd[13143]: Accepted password for root from 66.96.233.192 port 34454 ssh2 Jan 17 00:01:24 usaxen05 sshd[13143]: pam_unix(sshd:session): session opened for user root by (uid=0) Jan 17 00:01:25 usaxen05 sshd[13143]: pam_unix(sshd:session): session closed for user root Jan 17 00:01:26 usaxen05 sshd[13150]: Accepted password for root from 66.96.233.192 port 34455 ssh2 Jan 17 00:01:26 usaxen05 sshd[13150]: pam_unix(sshd:session): session opened for user root by (uid=0) Jan 17 00:01:27 usaxen05 sshd[13150]: pam_unix(sshd:session): session closed for user root Jan 17 00:01:43 usaxen05 sshd[13157]: Accepted password for root from 66.96.233.192 port 34456 ssh2 Jan 17 00:01:43 usaxen05 sshd[13157]: pam_unix(sshd:session): session opened for user root by (uid=0) Jan 17 00:01:44 usaxen05 sshd[13157]: pam_unix(sshd:session): session closed for user root Jan 17 00:01:45 usaxen05 sshd[13164]: Accepted password for root from 66.96.233.192 port 34457 ssh2 Jan 17 00:01:45 usaxen05 sshd[13164]: pam_unix(sshd:session): session opened for user root by (uid=0) Jan 17 00:01:46 usaxen05 sshd[13164]: pam_unix(sshd:session): session closed for user root

And there's no /var/log/authlog on my CentOS 5.4 server, so I can't check it. /var/log/messages also doesn't indicate anything strange.

Submitted by JamieCameron on Sat, 01/16/2010 - 16:47

That looks OK to me as well..

How much RAM does this host system have? It is possible that lack of RAM is causing un-expected failures like this ..

Submitted by SoftDux on Sat, 01/16/2010 - 17:00

As matter of interest, this server has 700MB RAM, according to "free -m" & "top"

root@usaxen05:[~]$ free -m total used free shared buffers cached Mem: 703 660 42 0 25 419 -/+ buffers/cache: 216 486 Swap: 3999 0 3999

top - 00:50:34 up 1 day, 1:29, 3 users, load average: 0.47, 0.24, 0.14 Tasks: 159 total, 3 running, 155 sleeping, 0 stopped, 1 zombie Cpu(s): 6.1%us, 3.7%sy, 0.0%ni, 88.5%id, 1.4%wa, 0.0%hi, 0.0%si, 0.3%st Mem: 719872k total, 712140k used, 7732k free, 26356k buffers Swap: 4095992k total, 172k used, 4095820k free, 429196k cached

so, I ran top while trying to create the VPS, and the load didn't jump much at the very moment that it should have created the LVM volume:

top - 00:57:58 up 1 day, 1:36, 3 users, load average: 0.26, 0.25, 0.18 Tasks: 154 total, 1 running, 153 sleeping, 0 stopped, 0 zombie Cpu(s): 0.1%us, 0.1%sy, 0.0%ni, 99.8%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st Mem: 719872k total, 65720k used, 24152k free, 19448k buffers Swap: 4095992k total, 176k used, 4095816k free, 477388k cached

So, I doubt if the amount of memory plays a role. The servr has 2GB total RAM, core2Quad 1.5Ghz CPU and no dom0 memory limits in the grub.conf file.

Submitted by JamieCameron on Sat, 01/16/2010 - 19:32

There must be something else going wrong then that we haven't seen before...

Any chance I could login myself to your Cloudmin master system and see what is really happening?