Hi, everybody, However, I made the following entry into initXXXX.ora file: Event=" 10046 trace name errorstack level 12" and it rebooted over the weekend. Then I got a call from my boss on Monday morning (I was home sick) letting me know the system had come to a crawl and was producing a trace file every minute or so. Eventually, after some speculations, I told him to rem out that entry and reboot. That resolved the issue, but left me puzzled. If anyone got any insight, it is appreciated, as always. Best, Sergey P.S. Trace files have the same header (different SIDs) and sizes between .5M and 3M Dump file c:\oracle\ora81\UDUMP\ORA00144.TRC Mon Jan 28 07:35:16 2002 ORACLE V8.1.7.0.0 - Production vsnsta=0 vsnsql=e vsnxtr=3 Windows 2000 Version 5.0 Service Pack 1, CPU type 586 Oracle8i Enterprise Edition Release 8.1.7.0.0 - Production With the Partitioning option JServer Release 8.1.7.0.0 - Production Windows 2000 Version 5.0 Service Pack 1, CPU type 586 Instance name: prod
Redo thread mounted by this instance: 1 Oracle process number: 10 Windows thread id: 144, image: ORACLE.EXE *** SESSION ID:(13.115) 2002-01-28 07:35:16.703 *** 2002-01-28 07:35:16.703 ksedmp: internal or fatal error ......ETC...... -----Original Message----- Sent: Monday, January 28, 2002 10:06 AM To: Multiple recipients of list ORACLE-L Subject: RE: Session_wait You're most welcome! Sounds like you've dug well into it and identified problem areas. ( Note John's comments below, too. Some echo my own, and other point to deeper issues in the SQL. ) You might be able to use this info to slap the vendor around in a public forum, creating embarassment and rapid system evolution. Yours in Slapping Developers Around, ross -----Original Message----- Sent: Friday, January 25, 2002 1:32 PM To: Multiple recipients of list ORACLE-L Hi, Ross, This has been running for about 4 hours now. I got an SQL trace file, and looked at the execution plans. So here's the deal. There's an interesting join condition "...where A.col1=B.col1.....". However, A has a few hundred distinct values in that column, none of them being NULL, and B, which has a few hundred thousand rows, has ALL NULLs in the corresponding column, and that column is not indexed, too. That's the query where it sits for a couple of hours. Guess what the optimizer is doing (8i)? I think internal effects are secondary in this scenario. It is the production (including the database) designed by the company named DELTEK, so nobody can change the code. Anyway, I reported my findings... Thank you very much for your help, it's always appreciated. Best, Sergey -----Original Message----- Sent: Friday, January 25, 2002 12:56 PM To: Multiple recipients of list ORACLE-L Subject: RE: Session_wait This is a busy little beaver...how long you say this runs? opened cursors cumulative 1072 session logical reads 77233609 db block gets 2642119 consistent gets 74591490 physical reads 148822 db block changes 3005410 consistent changes 141 no work - consistent read gets 72058049 table scans (short tables) 210918 table scans (long tables) 36 table scan rows gotten 798264962 table scan blocks gotten 71788386 table fetch by rowid 1074164 leaf node splits 4018 execute count 74445 bytes sent via SQL*Net to client 59650 bytes received via SQL*Net from client 84233 SQL*Net roundtrips to/from client 342 buffer is not pinned count 73513922 I see some updates (or inserts) and heavy reads...at first glance, though, you don't appear to be I/O throttled but likely have inefficiencies in buffer cache. Pin some small tables? Reexamine access paths for fts, even if on "small" tables (generally defined to be around 5% in blocks of buffer cache size?)and consider seeking out and destroying nested loops joins in favor of hash joins. There's more CPU, but less buffer cache splashing around....don't forget to review init.ora settings for hash, buffer pools, and query planning. (...be happy to do a flyover of those as well, if you like.) That's about all i can get in a one minute glance, but there are alot of people on the list who'll see more. Look to them for longer posts with more explanation. Good Luck! - Ross -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Mohan, Ross INET: [EMAIL PROTECTED] Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051 San Diego, California -- Public Internet access / Mailing Lists -------------------------------------------------------------------- To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Babich , Sergey INET: [EMAIL PROTECTED] Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051 San Diego, California -- Public Internet access / Mailing Lists -------------------------------------------------------------------- To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Mohan, Ross INET: [EMAIL PROTECTED] Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051 San Diego, California -- Public Internet access / Mailing Lists -------------------------------------------------------------------- To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). -- Please see the official ORACLE-L FAQ: http://www.orafaq.com -- Author: Babich , Sergey INET: [EMAIL PROTECTED] Fat City Network Services -- (858) 538-5051 FAX: (858) 538-5051 San Diego, California -- Public Internet access / Mailing Lists -------------------------------------------------------------------- To REMOVE yourself from this mailing list, send an E-Mail message to: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
