actually, the major problems are: Once workers > 1, squid looks for /var/run/squid. pidfile for workers=1 is done in squid.conf, but for workers > 1, this will be ignored.
once configuring localstatedir=/opt/deploy/squid/var using ./configure, for workers > 1, the squid is looking for /opt/deploy/squid/var/runsquid. The directory has to be created properly before starting squid. all helpers must be defined in process_number > 1 respectively, otherwise, there are extra helpers launched by process_number = 0 like crtd. The document is not clear on this, but we figure it out now. Alex From: [email protected] To: [email protected]; [email protected] Date: Fri, 10 Jul 2015 14:21:09 -0700 Subject: Re: [squid-users] sslbump and caching of generated cert figured it out. It needs all helpers under the process id. I have content redirect helpers which are not under the process ids. Alex > Date: Fri, 10 Jul 2015 15:44:49 +1200 > From: [email protected] > To: [email protected]; [email protected] > Subject: Re: [squid-users] sslbump and caching of generated cert > > On 10/07/2015 7:03 a.m., Alex Wu wrote: > > It seems the option http_port cannot be put under each process ID. If using > > workers, http_port cannot bind to ports specified from http_port. > > > > ?? Works for me and many others. > > What I dont expect to work is the *same* port line in two specific > workers unless the coordinator process also is assigned to mediate its use. > > The coordinator always needs a generic port for its own use. 3128 is > officially registered for that. But that is a separate problem to what > you are describing. > > Amos > _______________________________________________ squid-users mailing list [email protected] http://lists.squid-cache.org/listinfo/squid-users
_______________________________________________ squid-users mailing list [email protected] http://lists.squid-cache.org/listinfo/squid-users
