I saw the same thing as well when we first started up the cache server with
empty cache dirs. It would continue for several days and then stop.
During our testing we cleaned out the cache dirs several times and the
spiking
would restart.
Now it's be up over a month and we haven't had any more spikes since the
first few days.
We have no neighbours installed as well.
>
>
> We see the same problem. We're running 2.1.patch2. It has not
> been as bad since we increased the cache from 3 to 10 gb. But it
> still happens. This has occured without any neighbors configured.
> I, too, would like to know what is happening.
>
>
> The graphs can be seen at http://stats.bit-net.com/mrtg/reflect1.1.html
>
> > Gideon Glass writes:
> >
> > >During a spike, what seems to be happening is that Squid downloads one
> > >or more large HTTP objects very rapidly. In one ~2 second interval,
> > >tcpdump showed ~380KB on one TCP connection from a remote server to
> > >squid. In the same interval a single dialin client got ~15KB on its
> > >connection from the cache.
> >
> > Could it be a cache digest? Does this cache have any neighbors
> > configured?
> >
> > Duane W.
> >
> >
>
>
Received on Fri Jan 29 1999 - 12:41:50 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:44:11 MST