[squid-users] cachmgr output with 2.3.s4 and solaris 2.6 - something wrong with inodes?

as you can see in the cachemgr output below, the filesystem space in use
is much bigger than the actual size of the store directory.
the store swap size does not grow any longer as the limit is reached - so
squid
recognizes the cache limit parameters correctly.

subsequently, i guess this is a problem specific to solaris and its crappy
fs.

has anyone successfully fixed this yet?

-------------------------------------------------------------------------------------------

Store Directory Statistics:
Store Entries : 34295
Maximum Swap Size : 327680 KB
Current Store Swap Size: 304064 KB
Current Capacity : 93% used, 7% free

Store Directory #0: /cache1
First level subdirectories: 32
Second level subdirectories: 256
Maximum Size: 163840 KB
Current Size: 151950 KB
Percent Used: 92.74%
Filemap bits in use: 16852 of 32768 (51%)
Filesystem Space in use: 306547/6484674 KB (5%)
Filesystem Inodes in use: 40988/780160 (5%)
Flags: SELECTED

Store Directory #1: /cache2
First level subdirectories: 32
Second level subdirectories: 256
Maximum Size: 163840 KB
Current Size: 152114 KB
Percent Used: 92.84%
Filemap bits in use: 16695 of 32768 (51%)
Filesystem Space in use: 311967/6484674 KB (5%)
Filesystem Inodes in use: 39728/780160 (5%)
Flags:
Received on Thu May 31 2001 - 07:41:51 MDT