1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
|
[AssetCache]
;;
;; Options for FlotsamAssetCache
;;
; cache directory can be shared by multiple instances
CacheDirectory = ${Paths|CachePath}/assetcache
; Other examples:
;CacheDirectory = /directory/writable/by/OpenSim/instance
; Log level
; 0 - (Error) Errors only
; 1 - (Info) Hit Rate Stats + Level 0
; 2 - (Debug) Cache Activity (Reads/Writes) + Level 1
;
LogLevel = 0
; How often should hit rates be displayed (given in AssetRequests)
; 0 to disable
HitRateDisplay = 100
; Set to false for no memory cache
; assets can be requested several times in short periods
; so even a small memory cache is useful
MemoryCacheEnabled = false
; If a memory cache hit happens, or the asset is still in memory
; due to other causes, update the timestamp on the disk file anyway.
; Don't turn this on unless you share your asset cache between simulators
; AND use an external process, e.g. cron job, to clean it up.
UpdateFileTimeOnCacheHit = false
; Enabling this will cache negative fetches. If an asset is negative-cached
; it will not be re-requested from the asset server again for a while.
; Generally, this is a good thing.
;
; Regular expiration settings (non-sliding) mean that the asset will be
; retried after the time has expired. Sliding expiration means that
; the time the negative cache will keep the asset is refreshed each
; time a fetch is attempted. Use sliding expiration if you have rogue
; scripts hammering the asset server with requests for nonexistent
; assets.
;
; There are two cases where negative caching may cause issues:
;
; 1 - If an invalid asset is repeatedly requested by a script and that asset is
; subsequently created, it will not be seen until fcache clear
; is used. This is a very theoretical scenario since UUID collisions
; are deemed to be not occuring in practice.
; This can only become an issue with sliding expiration time.
;
; 2 - If the asset service is clustered, an asset may not have propagated
; to all cluster members when it is first attempted to fetch it.
; This may theoretically occur with networked vendor systems and
; would lead to an asset not found message. However, after the
; expiration time has elapsed, the asset will then be fetchable.
;
; The defaults below are suitable for all small to medium installations
; including grids.
NegativeCacheEnabled = true
NegativeCacheTimeout = 120
NegativeCacheSliding = false
; Set to false for no file cache
FileCacheEnabled = true
; How long {in hours} to keep assets cached in memory, .5 == 30 minutes
; even a few minutes may mean many assets loaded to memory, if not all.
; this is good if memory is not a problem.
; if memory is a problem then a few seconds may actually save same.
; see hit rates with console comand: fcache status
MemoryCacheTimeout = .016 ; one minute
; How long {in hours} to keep assets cached on disk, .5 == 30 minutes
; Specify 0 if you do not want your disk cache to expire
FileCacheTimeout = 48
; How often {in hours} should the disk be checked for expired files
; Specify 0 to disable expiration checking
FileCleanupTimer = 0.0 ; disabled
; If WAIT_ON_INPROGRESS_REQUESTS has been defined then this specifies how
; long (in miliseconds) to block a request thread while trying to complete
; an existing write to disk.
; NOTE: THIS PARAMETER IS NOT CURRENTLY USED BY THE CACHE
; WaitOnInprogressTimeout = 3000
; Number of tiers to use for cache directories (current valid
; range 1 to 3)
;CacheDirectoryTiers = 1
; Number of letters per path tier, 1 will create 16 directories
; per tier, 2 - 256, 3 - 4096 and 4 - 65K
;CacheDirectoryTierLength = 3
; Warning level for cache directory size
;CacheWarnAt = 30000
|
