Question one: Failed source handling
I have 3 Vhosts, each with 2 Python backend sources for load
balancing. I noticed that when 1 source misbehaves, and Cherokee emits
errors like this:
{'type': "warning", 'time': "27/04/2011 09:16:47.358", 'title':
"Sources exhausted: re-enabling one.", 'code': "balancer_round_robin.c:
122", 'error': "64", 'description': "All the Information Sources have
been off-lined. ....
It takes the whole server down, which is terrible. Every source lives
in its own environment, why should 1 take down 3 domains, the default
host and 5 other innocent sources?
How can I configure Cherokee to continue to "live" even though a
source is misbehaving?
Question two: Cherokee admin
I noticed, for several versions back now, that Cherokee-Admin does not
consequently detect when Cherokee is up or down. It is very bothersome
that Upstart on Ubuntu 10.04 64 bit brings Cherokee automatically back
on and the status is not detected by the admin. I make my changes, try
to restart and get errors that another server is using port 80. I have
to stop Cherokke admin, kill the newly started Cherokee worker
process, restart Cherokee-Admin just to save my changes and restart.
Is there a way to set the status polling to a higher value in Cherokee-
Admin?
Question 3: Wildcard matching
I would like to match a domain like this "http://mydomain.com"; or
"www.mydomain.com". I put "*.mydomain.com" as the wildcard value, this
works but does not match "http://mydomain.com";.
Can someone tell me what I am doing wrong?
My system is an Ubuntu 10.04 64bit server with the latest Cherokee
Thanks
_______________________________________________
Cherokee mailing list
[email protected]
http://lists.octality.com/listinfo/cherokee
