On Tue, Aug 18, 1998 at 10:51:43AM -0600, Duane Wessels wrote:
> Any write errors in your log file?
Only one... and it doesn't make sense. As best as I can tell, the
disks _never_ reached capacity, and besides, if you look at the
cachemgr dump, the Total Swap Size is off - and it looks like this is
what is causing the LRU mecahnism to drop stuff off the end...
If you look at the stats I just got:
Store Directory Statistics:
Store Entries : 761325
Maximum Swap Size : 8560752 KB
Current Store Swap Size: 7817328 KB
Current Capacity : 91% used, 9% free
Store Directory #0: /mounts/cache0/cache
First level subdirectories: 64
Second level subdirectories: 256
Maximum Size: 3932160 KB
Current Size: 3067778 KB
Percent Used: 78.02%
Filemap bits in use: 296474 of 655360 (45%)
Store Directory #1: /mounts/cache1/cache
First level subdirectories: 64
Second level subdirectories: 256
Maximum Size: 3932160 KB
Current Size: 3067880 KB
Percent Used: 78.02%
Filemap bits in use: 304805 of 655360 (47%)
Store Directory #2: /mounts/cache2/cache
First level subdirectories: 64
Second level subdirectories: 256
Maximum Size: 3932160 KB
Current Size: 1681670 KB
Percent Used: 42.77%
Filemap bits in use: 159959 of 655360 (24%)
You'll see none of the three swap spaces are near 90%, but
collectively they all are because the Total Size is wrong.
I tried to work out how it was miscalculating the total size, but
can't seem to come up with anything that makes sense.
-cw
Received on Tue Jul 29 2003 - 13:15:51 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:11:51 MST