Author: mocha Email: [EMAIL PROTECTED] Message: NetBSD/Alpha 1.5.1, mnogosearch-3.1.8, pgsql-7.1-cvs i\'m indexing all servers in (just the important params): DeleteNoServer no Follow world Server world http://www.state.ca.us/ Realm ^http://.*\\.state\\.ca\\.us/ first let me say that i think there is a bug in the code to check for network time-outs. i\'m currently trying to index all the regional servers in my state. therefore, the latency time could vary from server to server. i don\'t think you\'ll find indexers staling in idle if you are indexing local servers on your intranet. i would start about 10 indexers at night and let it run overnight. in the morning i would find all of the processes stalling in idle mode. there weren\'t any errors and nothing in the log files. i\'ve described this problem in the past few days, so it\'s archived. after trying all kind of different things, including vacuum analyze the db a few times, nothing seemed to help. though the sleep time parameter of 2 seconds or more did seem to help, but eventually the indexer process would hang. it varied from 1 min to several hours. i finally chased the problem down to these settings. this is what i had in my indexer.conf originally when indexers were stalling: MaxHops 256 MaxNetErrors 16 ReadTimeOut 1m30s after i changed those settings to: MaxHops 50 MaxNetErrors 6 ReadTimeOut 30s i no longer see any indexer processes hang. however, occasionally i do see something that i didn\'t see before: Indexer[28982]: [1] Download timeout that was what led me to believe it was the network timeout issue. but at this time, i\'d rather get the \"Download timeout\" than having indexer stall and not index any other sites. Reply: <http://search.mnogo.ru/board/message.php?id=1024> ______________ If you want to unsubscribe send "unsubscribe udmsearch" to [EMAIL PROTECTED]
