Clicking through to the description of the poll_timeout setting gives you a better description of what is going on. https://docs.trafficserver.apache.org/en/latest/admin-guide/files/records.config.en.html#proxy-config-net-poll-timeout
If your poll is set too low, your threads will spend very little time blocking in the poll call, and most of the time spinning between poll calls. How are you measuring your CPU load? Total %CPU over all the CPUs on your system (e.g. %CPU at the top of your "top" display) or %CPU per computation unit (e.g. %CPU for your traffic_server process in "top"). On an idle Centos7 box, I see top's reporting of CPU utilization for traffic_server reported by top at 7.6% but total machine CPU at 0.2% On Wed, Dec 18, 2019 at 3:37 AM Veiko Kukk <[email protected]> wrote: > Hi > > Before I try anything, I need to understand what it means. > What does this option do? What polling? net indicates it has something > to do with network, but what does it poll on network? > > Veiko > > On Sun, 15 Dec 2019 at 20:25, Shu Kit Chan <[email protected]> wrote: > > > > Perhaps try this out? - > > > https://docs.trafficserver.apache.org/en/latest/admin-guide/performance/index.en.html#polling-timeout > > > > On Fri, Dec 13, 2019 at 6:07 AM Veiko Kukk <[email protected]> wrote: > > > > > > Hi > > > > > > ATS 7.1.2, Centos 7.7. > > > I'm seeing 15-16% CPU load while there is no traffic at all. > > > proxy.process.cache.bytes_used 23934931615744 > > > proxy.process.cache.bytes_total 23935065833472 > > > > > > Is that normal? > > > > > > Veiko >
