From: Sambavan Newsgroups: comp.databases.oracle.server Subject: MTS server process seems to die and resume at random Date: Tue, 10 Oct 2000 05:23:13 GMT Organization: Deja.com - Before you buy. Lines: 68 Message-ID: <8ru93t$br4$1@nnrp1.deja.com> NNTP-Posting-Host: 129.116.78.131 X-Article-Creation-Date: Tue Oct 10 05:23:13 2000 GMT X-Http-User-Agent: Mozilla/4.74 [en] (WinNT; U) X-Http-Proxy: 1.0 x72.deja.com:80 (Squid/1.1.22) for client 128.83.93.51, 129.116.78.131 X-MyDeja-Info: XMYDJUIDsambavan Folks, I'm running my oracle 8.1.6 STD on solaris8, sun4u. Lately I have been getting this ora-600 which seems to get cause by the crusade of my MTS. I couldn't find the crusader :P Here is a log that got generated by the alert log ARC0: Completed archiving log# 1 seq# 23760 Mon Oct 9 22:02:25 2000 Errors in file /usr/local/oracle/app/oracle/admin/active/bdump/active_s000_21858.trc: ORA-00600: internal error code, arguments: [16365], [2323556496], [1], [4], [], [], [], [] Mon Oct 9 22:02:29 2000 Errors in file /usr/local/oracle/app/oracle/admin/active/bdump/active_s001_28852.trc: ORA-00600: internal error code, arguments: [16365], [2323564176], [0], [4], [], [], [], [] Mon Oct 9 22:02:41 2000 Thread 1 advanced to log sequence 23762 Mon Oct 9 22:02:41 2000 ARC2: Beginning to archive log# 5 seq# 23761 Mon Oct 9 22:02:41 2000 Current log# 2 seq# 23762 mem# 0: /usr/local/oracle/app/oracle/oradata/active/redo02.log Mon Oct 9 22:02:42 2000 ARC2: Completed archiving log# 5 seq# 23761 Mon Oct 9 22:03:23 2000 found dead multi-threaded server 'S000', pid = (12, 3) found dead multi-threaded server 'S001', pid = (23, 34) Mon Oct 9 22:05:42 2000 Thread 1 advanced to log sequence 23763 Mon Oct 9 22:05:42 2000 ARC1: Beginning to archive log# 2 seq# 23762 The process seems to have resumed after that dump. But have no clue on what is causing this. For your reference, here is some stuff from my trace file Dumping MTS state at level 10 Global: Svc Name=active, Max Svrs=20, Max Disp=5, Max Ses=110 Svrs: #act=1, #sta=49, #trm=75, #unt=26, #max=9 Starting: flg=20, current=1, index=7 In this dump I could notice that there were around six server porcess that were free at the time of dump. So, this had puzzled me further on finding out what was causing the problem. On this instance,I use the php to open persistent connection to the database one user. But under that user profile, I have also setted the idle time to be max of 5min. So would this connection be disconnected, if the session was idle for 5min or would it be ignored? Could this be a cause for this problem? Is there any tracking that I could do on this to see what is causing the problem? Since this problem seems to be random, does not know what could be the causing it, I have no clue on what the steps/measures to take to troubleshoot it. Some help/thoughts would really be appreciated. Thanx for your time and thoughts, Sambavan Sent via Deja.com http://www.deja.com/ Before you buy.