[squid-users] squid SMP notes

senor frio_cervesa at hotmail.com
Thu Dec 8 08:07:26 UTC 2016


On 12/7/2016 13:43, Eliezer Croitoru wrote:
> May I ask about the purpose of the proxy? Caching or ACL?
> 
> Eliezer
> 
> ----
> http://ngtech.co.il/lmgtfy/
> Linux System Administrator
> Mobile: +972-5-28704261
> Email: eliezer at ngtech.co.il
> 
> 
> From: squid-users [mailto:squid-users-bounces at lists.squid-cache.org] On
> Behalf Of senor
> Sent: Thursday, December 1, 2016 8:11 AM
> To: squid-users at lists.squid-cache.org
> Subject: [squid-users] squid SMP notes
> 
> It is unclear to me how the more recent 'multiple instance' configuration
> compares with that of 'SMP Scaling'. 
> 
> Is there a short and sweet description of pros and cons or is the multiple
> instance simply the new way? The wiki pages vary in detail of implementation
> but don't directly state any advantage of one over the other. Further
> confusing me is my need to consider SSL Interception, ssl_crtd helpers and
> ecap. To be clear, I'm not looking for configuration help. Just some advice
> on which to pursue so I'm not rethinking the whole thing 6 months down the
> road.
> Thank you for any help.
> 
> 
A combination of ACL, eCAP and caching is my goal. At the moment I'm
trying to make sense out of the statistics so I have metrics to guide
me. So now it's just a matter of comparing the various possible
configurations.

Is there a tool or option to log statistics in machine readable format?
SNMP doesn't expose as much as I'd like and cachemgr output needs to be
gawked down to something like a csv file but still account for kids?

Also, is ssl_crtd disk DB ok to share among workers or do they each need
their own? I just realized I forgot to give the workers their own DB but
it seems to be working. Although not under load.

I really appreciate the help.


More information about the squid-users mailing list