Submitted by chriswik on Wed, 07/01/2015 - 16:56 Pro Licensee
I have a 4GB system image cache configured on all our Xen hosts.
However Cloudmin won't use it anymore, I'm not exactly sure when this started but I think one of the recent updates, as it used to work. Now provisioning a VM on a Xen host in another datacenter takes hours, or just times out, as it tries to pipe the system image over ssh and dd to the lvm lv.
Can you fix it please?
Status:
Closed (fixed)
Comments
Submitted by JamieCameron on Thu, 07/02/2015 - 00:53 Comment #1
This could be due to a change to reduce temporary disk space usage on the host by writing to the new VM's disk directly via SSH.
How distant is your Cloudmin master system from the Xen host?
Submitted by chriswik on Thu, 07/02/2015 - 03:13 Pro Licensee Comment #2
Cloudmin master is in Amsterdam, we also have Xen deployments in Chicago and Singapore.
Singapore is the most troublesome, the others work OK.
Submitted by JamieCameron on Thu, 07/02/2015 - 11:42 Comment #3
If you go to Cloudmin Setings -> CLoudmin Configuration, in the "Virtual system settings" section you can change the "Extract images directly from the Cloudmin master?" option to "No", which will force use of the local cache on the host.
Submitted by chriswik on Thu, 07/02/2015 - 12:02 Pro Licensee Comment #4
Perfect, thanks!
Amsterdam <-> Singapore latency can be killer depending on time of day. Singapore <-> USA is actually pretty good...
Submitted by JamieCameron on Thu, 07/02/2015 - 12:26 Comment #5
That's because most network connections take the long way around the world :-)
Submitted by Issues on Thu, 07/16/2015 - 12:29 Comment #6
Automatically closed -- issue fixed for 2 weeks with no activity.
Submitted by Issues on Thu, 07/16/2015 - 12:29 Comment #7
Automatically closed -- issue fixed for 2 weeks with no activity.
Submitted by Issues on Thu, 07/16/2015 - 12:29 Comment #8
Automatically closed -- issue fixed for 2 weeks with no activity.