About the FD limit, it's pretty safe to use a big number of FD limit such as 65k(65535) for a period of time and if it works fine then leave it as is. Is Jessie using a systemd unit file? If it's using a system unit try to add into the service file under the service section: [Service] LimitNOFILE=65535 And restart the service. Unless you will update squid it should stay there for a long time. Then after you will identify if it works for you or not you can copy the systemd unit service file into: /etc/systemd/system And add save this change in it. Then it should override the default: /usr/lib/systemd/system/<servicename>.service (as far as I remember) Let us know if it works for you, Eliezer ---- Eliezer Croitoru Linux System Administrator Mobile: +972-5-28704261 Email: eliezer@xxxxxxxxxxxx -----Original Message----- From: squid-users [mailto:squid-users-bounces@xxxxxxxxxxxxxxxxxxxxx] On Behalf Of Amos Jeffries Sent: Wednesday, August 23, 2017 22:58 To: squid-users@xxxxxxxxxxxxxxxxxxxxx Subject: Re: Squid stopped writing the log files... just for a while. On 24/08/17 01:42, Antony Stone wrote: > On Wednesday 23 August 2017 at 15:26:53, Verónica Ovando wrote: > >> This is the last request until it holdups: >> 1503416750.346 386 192.168.0.2 TCP_REFRESH_UNMODIFIED/304 235 GET >> http://www.mysite.com user HIER_DIRECT/www.mysite.com - This is the next >> request: >> 1503416781.864 141376 192.168.0.2 TCP_MISS/200 5599 CONNECT >> play.google.com:443 user HIER_DIRECT/play.google.com - > > Okay, so I see there's a 36 second gap between those two requests; were > clients sending requests during that time? > Ah, the second log entry is for a CONNECT tunnel that was started 111.5s before the first request and took 140 seconds to deliver all its 5KB of tunneled traffic. The time it was open and not doing anything useful might be related to the delays, but not an indicator of anything specifically related. It could just be a client holding an HTTPS connection with keep-alive until the TCP layer or Squids timeouts kill it. (The available FD slots running out temporarily, or bursts of socket FDs going through TCP *_WAIT states could be the reason for pausing). Amos _______________________________________________ squid-users mailing list squid-users@xxxxxxxxxxxxxxxxxxxxx http://lists.squid-cache.org/listinfo/squid-users _______________________________________________ squid-users mailing list squid-users@xxxxxxxxxxxxxxxxxxxxx http://lists.squid-cache.org/listinfo/squid-users
