On Sun, 4 Mar 2001, Artem Veremey wrote:
> >>>> == chin60@pchome.com.tw <chin60@pchome.com.tw> writes:
> > What is the normal memory usage of polyclt? If the
> > polyclt's memory usage is increasing. Does it mean that
> > the requests in wait queue is increasing? Does it mean that
> > our peak request rat be set to high? Thanks!
>
> If polyclt memory usage increases over time then there is memory
> leak (which is very unlikely). It has nothing to do with requests
> waiting in queue.
There are no known memory leaks in 2.5.4. Requests that are waiting
for connection slot to become available (see open_conn_lmt) do consume
memory and will increase memory footprint if the queue is growing. If
the queue is growing, Polygraph may run out of memory (sooner or
later, depending on the queue growth speed and available memory).
Finally, as I tried to tell chin60 several times, if cache response
time is too high for a given workload, it probably means that the
request rate is too high for that cache configuration, and that
polyclt running PolyMix workload may eventually run our of memory
under these conditions. This behavior is intentional to prevent
best-effort "PolyMix" tests (which PolyMix becomes if the queues are
not drained) from being considered successful.
Polygraph 2.6 will attempt to flush the logs when it runs out of
memory. Earlier versions will certainly loose phase statistics in
binary logs when virtual memory is exhausted.
Alex.
This archive was generated by hypermail 2b29 : Tue Jul 10 2001 - 12:00:18 MDT