Hi,
I have struggled quite a bit to get it up and running. Over the time, I have accumulated
some notes on various things I did so to share with everyone who is interested in doing
this. This complements the online doc in a way that might give me a complete picture in
one place. However I need some clarifications as I might have forgotten to document
certain steps or certain steps I did turn out to be not necessary in the end. It will be
great if experts here can help me get the things straight.
My setup is like:
Browser (firefox 24.2 on RHEL6) ------------ ovirt-engine (3.3.2) ------------ ovirt-node
(3.0.3)
No direct network connectivity from the browser machine to the node machine.
These are the major things I installed for spice-proxy to work:
* On ovirt-engine:
yum install spice-gtk, virt-viewer, spice-xpi
yum-install squid
/etc/squid/squid.conf updates:
acl localhost src <browser IP addr>
#http_access deny CONNECT !SSL_ports
http_access deny !Safe_ports
http_port 3128
service squid restart
make sure iptables allow 3128
engine-config -s SpiceProxyDefault= http://<ovirt-engine-IP>:3128
service ovirt-engine restart
* On browser machine running firefox 24.2.0 on RHEL6 for running browser console
plugin client
yum install spice-xpi.
make sure VM's console option is set to SPICE
Are the above steps reasonable? any missing or redundant?
Additional questions:
1. Will spice-proxy work with the Spice HTML5 client in the browser?
2. Is the spice-proxy architecture diagram like: browser --------- squid proxy -
spice-proxy ---------------------- VM
3. I didn't explicitly install any certs for the squid proxy. Is it automatically
taken care of?
References:
http://www.ovirt.org/Console_Client_Resources
http://www.ovirt.org/Features/Spice_Proxy
https://access.redhat.com/site/documentation/en-US/Red_Hat_Enterprise_Vir...
Thanks.
David