On 24/06/11 18:24, Jenny Lee wrote:
>>
>> It would be nice to have something like a dashboard feature for squid (squid
>> \
>> serving it):
>> https://www.varnish-software.com/sites/default/files/vac_dashboard_15062011.png
>>
>> It shouldn't be too difficult to implement I believe since these values are \
>> available in cachemgr.
>> Unfortunately, many of us don't have web servers where cachemgr is installed.
>>
>> Even if we had, it is not easy to understand what is what when all we care
>> about is \
>> few simple metrics (reqs served, avg traffic in/out, reqs/sec, total
>> traffic, etc). \
>> Jenny
> Thank you for pointing that out. It has been on the wishlist for a while
> with very little interest.
I think this would be a cool feature without being dependant on external tools.
> There are a few small tools out there which use the cachemgr or SNMP
> interface to Squid and generate that kind of report live or graphical
> forms. They just tend to get lost in the flood of log analysis tools.
> Cacti, MySAR, and Duane's old rrd scripts etc.
Yes I know, Amos. I have 800 of those tools. I use monitorix and sometimes
nagios to pull these. But after a while, when servers stack up and installs
increase, OS'es multiply, you tend to lose track of what is what, or what is
where.
Besides, cachemgr is rather cryptic for us mere mortals. I am looking at this
delay-pools section in cachemgr last 10 years and I still don't understand what
it represents or what is happenning with the guy I put in the pool. Something
more user-friendly with little explanations would be nice.
squid
-->log_file_deamon
-->ncsa_auth /squid/members
-->reports-console (listening on 3129, accesible to localhost or whatever
allowed in squid.conf)
Jenny
PS: Please cc:, not on the list.
