Reducing overflowed and dropped work requests

Simon Males sime at sime.net.au
Fri Jul 15 03:15:27 CEST 2011


> The Varnish munin plugins have alerted me a number of times (say less
> then 5) that N dropped work requests (n_wrk_drop) is greater then 1.
>
> According to Kristian Lyngstol's post [1] this is represents when the
> request queue it full. Additionally I believe the overflowed work
> requests (n_wrk_overflow) will always be greater then n_wrk_drop.
>
> I think I should be trying to tackle two issues: Limit requests
> overflowing, followed by increasing the size of the request queue.
>
> Of the below startup parameters I believe thread_pools should match
> the number of cores the server has and increase the maximum number of
> threads to possibly double (from 25 to 50?).
>
> DAEMON_OPTS="-a :80 \
>             -T localhost:6082 \
>             -f /etc/varnish/default.vcl \
>             -S /etc/varnish/secret \
>             -w 10,25,120 \
>             -p listen_depth=8192 \
>             -p log_hashstring=off \
>             -p lru_interval=60 \
>             -p sess_timeout=10 \
>             -p shm_workspace=32768 \
>             -p ping_interval=1 \
>             -p thread_pools=4 \
>             -s malloc,2G"

As overflow_max is default set to 100%, I believe I essentially need
to increase the number of worker threads to increase the ratio. To
increase the threads, I'll increase the thread_pools.

-- 
Simon Males




More information about the varnish-misc mailing list