[ https://issues.apache.org/jira/browse/TS-343?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Leif Hedstrom updated TS-343: ----------------------------- Fix Version/s: (was: 3.1.6) 3.3.0 Moving out to v3.3.0, move back to 3.1.4 if this will be work on *soon*. > The cache lookup and add operation use different key in plugin > -------------------------------------------------------------- > > Key: TS-343 > URL: https://issues.apache.org/jira/browse/TS-343 > Project: Traffic Server > Issue Type: Bug > Components: Cache > Affects Versions: 3.0.0 > Reporter: Flier Lu > Priority: Critical > Fix For: 3.3.0 > > > I'm developing a cache plugin base on the redis, > http://code.google.com/p/rediscache/ > The plugin could be loaded and hook the cache read/write operations > [May 9 16:25:05.012] Server {3079476928} NOTE: loading plugin > 'libexec/trafficserver/redis_cache.so' > [May 9 16:25:05.014] Server {3079476928} DIAG: (cache_plugin) > [INKPluginInit] Starting redis cache plugin > But the cache lookup and add operation use different key, it seems use the > url as lookup key > [May 9 16:25:13.149] Server {3066555280} DEBUG: (cache_plugin) > [CacheProcessor::open_read] Cache hooks are set > [May 9 16:25:13.153] Server {3066555280} DEBUG: (cache_plugin) > [NewCacheVC::alloc] new B33B1EE0 > [May 9 16:25:13.153] Server {3066555280} DEBUG: (cache_plugin) > [NewCacheVC::set_cache_http_hdr] > [May 9 16:25:13.153] Server {3066555280} DIAG: (cache_plugin) [cache_read] > [May 9 16:25:13.153] Server {3066555280} DEBUG: (cache_plugin) > [INKCacheKeyGet] vc get cache key > [May 9 16:25:13.158] Server {3066555280} DIAG: (cache_plugin) cache hitted > for key: http://www.baidu.com/ w/ 0 bytes value > [May 9 16:25:13.158] Server {3066555280} DEBUG: (cache_plugin) > [INKHttpCacheReenable] event id: 1133 data: 0 size: 0 > [May 9 16:25:13.158] Server {3066555280} DEBUG: (cache_plugin) > [INKHttpCacheReenable] cache_lookup_complete > [May 9 16:25:13.158] Server {3066555280} DEBUG: (cache_plugin) > [INKHttpCacheReenable] open read failed > but store the cache item with a random MD5 as key > [May 9 16:25:13.334] Server {3079476928} DEBUG: (cache_plugin) > [NewCacheVC::handleWrite] event=1 > [May 9 16:25:13.334] Server {3079476928} DIAG: (cache_plugin) [cache_write] > [May 9 16:25:13.334] Server {3079476928} DEBUG: (cache_plugin) > [INKCacheKeyGet] vc get cache key > [May 9 16:25:13.346] Server {3079476928} DIAG: (cache_plugin) put 3571 bytes > value to redis w/ 16 bytes key: 0xb33b1fb0 > [May 9 16:25:13.346] Server {3079476928} DEBUG: (cache_plugin) > [INKHttpCacheReenable] event id: 1129 data: 0 size: 3571 > [May 9 16:25:13.346] Server {3079476928} DEBUG: (cache_plugin) > [INKHttpCacheReenable] cache_write > I'm not sure whether it is a design issue or bug ? and the cache lookup > always has 0 size buffer ? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira