Hi *Squiders*,
it's me again with another confusing cache.log message ! :-)
When I start my squid, I see
97/01/23 10:15:03| file_map_create: creating space for 2097152 files
97/01/23 10:15:03| Swap maxSize 1572864, estimated 78643 objects
97/01/23 10:15:03| Target number of buckets: 1572
97/01/23 10:15:03| Using 7951 Store buckets, maintain 1 bucket every 10
seconds
It says estimated 78643 objects - but when cachemgr tells me the truth, I
now have 80178 StoreEntries. Is this bad? Will this be a reason for squid
to enter the purge mode? (if yes this would answer my previous mail where
I asked why my squid is in purge mode but there are about 200 MB left to
the high watermark)
Since I still have 500 MB left to my 1536 configured cache_swap, but the
estimated objects are already used, my average obj_size is to big ?
I did not change anything from this options when I installed my squid 1.1
the first time, so it must be usind the default 20 K which seems to big.
How is our experience with the default 20 k ? What value are you using, to
make your estimated object size to fit to your cache_swap size ?
And another question: Can I simply change the average obj_size value in my
squid.conf and then purge mode stops until high watermark is reached, or
does changing this value requires to build a new cache subdir structure?
(with loosing all cached data :-(((
Andreas
-- Andreas Reeh = A.Reeh@WaD.org * 51545 Waldbroel * Germany * PGP welcome More info, public key, geek-code and other email and talk addresses at http://www.wad.org/~andreas/ or http://home.pages.de/~andreas.reeh/Received on Thu Jan 23 1997 - 14:31:40 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:34:10 MST