Squid_ The Definitive Guide - Duane Wessels [150]
collisions on rej
This is the percentage of rejected additions that wouldn't have turned on any new bits.
storedir: Store Directory Stats
This page displays some statistics from the storage system. First, you'll see a few global values. For example:
Store Directory Statistics:
Store Entries : 2873564
Maximum Swap Size : 46080000 KB
Current Store Swap Size: 41461672 KB
Current Capacity : 90% used, 10% free
Store Entries
The number of StoreEntry objects. Most, but not necessarily all, of these are for on-disk objects.
Maximum Swap Size
The sum of all cache_dir sizes.
Current Store Swap Size
The total amount of cached data currently stored on disk. Note that Squid rounds response sizes (e.g., 1722 bytes) up to the nearest multiple filesystem block size (e.g., 2048 bytes) when incrementing and decrementing this value.
Current Capacity
The percentage of the maximum disk space currently in use. The percentage in use should normally stay below the cache_swap_high value.
Next, you'll see a section for each cache_dir. It looks something like this:
Store Directory #1 (diskd): /cache1
FS Block Size 1024 Bytes
First level subdirectories: 16
Second level subdirectories: 64
Maximum Size: 15360000 KB
Current Size: 13823996 KB
Percent Used: 90.00%
Filemap bits in use: 958439 of 2097152 (46%)
Filesystem Space in use: 14030485/17370434 KB (81%)
Filesystem Inodes in use: 959440/4340990 (22%)
Flags: SELECTED
Pending operations: 0
Removal policy: lru
LRU reference age: 23.63 days
Store Directory #
The directory number, type, and pathname.
FS Block Size
The filesystem block size, determined by the statfs( ) or statvfs( ) system calls. If these functions aren't available or return an error, the block size defaults to 2048 bytes.
The next few lines are actually storage scheme-dependent. For the most part, ufs, aufs, and diskd are very similar and all report the same statistics.
First level subdirectories
The number of first-level subdirectories you told Squid to use on the cache_dir line.
Second level subdirectories
The number of second-level subdirectories you told Squid to use on the cache_dir line.
Maximum Size
The maximum allowed size for this cache directory.
Current Size
The amount of disk space currently in use.
Percent Used
The percentage of cache_dir space currently in use.
Filemap bits in use
Squid uses a bitmap to keep track of file numbers that are allocated and free. This line shows the number and percentage of bits in use. The filemap grows automatically as needed, so don't worry if it shows up as 99% full.
Filesystem Space in use
These numbers come from the statfs( )/statvfs( ) system calls. These should be the same values as you'd see from the df command. Squid doesn't use these numbers, other than to report them here for your information. Note that these values may be larger than Current Size, especially if the partition is used for more than Squid's cache.
Filesystem Inodes in use
These numbers also come from statfs( )/statvfs( ). They are present to remind you that running out of inodes is just as bad as running out of free space. Unfortunately, if you run out of inodes, you'll probably be forced to newfs the partition.
Flags
Possible values include SELECTED and READ-ONLY. The SELECTED flag means that this particular cache_dir was most recently selected by the cache directory selection algorithm (see Section 7.4). The READ-ONLY flag means that the cache directory has been marked read-only in the configuration file (see Section 7.1.5).
Pending operations
This line appears only for diskd cache directories. It shows the number of I/O requests dispatched to the diskd process that have not yet been acknowledged.
That's the end of the scheme-specific data. The remaining lines are specific to the cache_dir replacement algorithm:
Removal policy
Possible values include lru (the default) or heap. Note that for heap,