Our cache has been hopping along nicely for quite a long time, doing up to 100req./sec. without many problems.
Since a few weeks, however, we repeatedly see the following scenario: When the general network load goes up (we have reserved soem bandwidth for the cache), first the number of requests/sec drops dramtically, then the reaction time increases steeply (up to more that a second for a *hit*). Unfortunately, this state is stable :-) Stopping and restarting squid helps. We have checked the rest of the load on the machine, but there's enough CPU left, there's enough memory left, and it isn't the listen queue either. I'm running out of ideas where to look. We are already planning a separate cache on a Linux system, but I would really like to figure this one out. Thanks for any help Jost -- | [EMAIL PROTECTED] Please help stamp out spam! | | Postmaster, JAPH, resident answer machine am RZ der RUB | | Pluralitas non est ponenda sine necessitate | | William of Ockham (1285-1347/49) |
