On Mon, Aug 22, 2011 at 8:47 PM, Hugo Cisneiros (Eitch) <[email protected]> wrote: > On Mon, Aug 22, 2011 at 7:27 PM, Hugo Cisneiros (Eitch) > <[email protected]> wrote: >> Any ideas? Maybe something hanging up the sessions (network latency >> between varnish and backend? or someone doing this on purpose?) > > Also, I got some errors in syslog regarding max memory maps per > process, found in this ticket: > > https://www.varnish-cache.org/trac/ticket/784 > > Since I raised the value from vm.max_map_count I don't get this error > anymore, but I'm seeing very high values up to 100k+... Is this > normal?
I upgraded the servers to Varnish 3.0 (with same configuration) and stopped getting Connection Reset errors. The memory consumption is still high, but not as high to give me errors. It's working fine now with high traffic for days along =) memory maps per proccess is a lot less to. 100k+ to 12k stable. The only strange thing is sometimes I'm still getting many stats on 'n_wrk_max - N worker threads limited', but it seems that the worker queue is working just fine. Thanks =) -- []'s Hugo www.devin.com.br _______________________________________________ varnish-misc mailing list [email protected] https://www.varnish-cache.org/lists/mailman/listinfo/varnish-misc
