RE: Extremly long checkpoints: How to find the reason and solve t

also, since there are lot of temp tables, make sure the temp tables are created with "with no log" option, if it is a logged database. -----Original Message----- From: owner-informix-list@iiug.org [mailto:owner-informix-list@iiug.org]On Behalf Of Savio Pinto (s) Sent: Tuesday, May 11, 2004 9:45 AM To: 'js@jetsys.de'; informix-list@iiug.org Subject: RE: Extremly long checkpoints: How to find the reason and solve t he problem (7.31 FD3) did the checkpoint suddenly went up, or is it normal in your environment?, it looks like you need to do some tuning, there are three area that i can think of that you need to tune 1) Hardware: make sure there are no i/o bottleneck on the disk, if you are using EMC make sure read and write cache is turned on. 2)Informix Engine: You may need to tune the following onconfig parameters LRU, LRU_MAX_DIRTY, LRU_MIN_DIRTY, CLEANERS, BUFFERS, CKPTINTVL, NUMAIOVPS, NUMCPUVPS, PHYSFILE. you can use the onstat commands to monitor the engine and see what needs to be tuned. also you may need to check your instance layout and the ddl's for the tables. 3) Application: make sure the program/logic/queries in the programs are efficient. at times even if the hardware and informix engine are configured efficiently, a bad program/queries/logic will cause lot of problems. -----Original Message----- From: owner-informix-list@iiug.org [mailto:owner-informix-list@iiug.org]On Behalf Of js@jetsys.de Sent: Tuesday, May 11, 2004 3:27 AM To: informix-list@iiug.org Subject: Extremly long checkpoints: How to find the reason and solve the problem (7.31 FD3) Hello, we have some serious problems with extremly long checkpoints. Interactive users can't really work. 09:01:15 Logical Log 28452 Complete 09:01:16 Logical Log 28452 - Backup Started 09:01:36 Logical Log 28452 - Backup Completed 09:30:28 Checkpoint Completed: duration was 1465 seconds. 09:30:28 Checkpoint loguniq 28453, logpos 0x2ac60 09:43:43 Checkpoint Completed: duration was 495 seconds. 09:43:43 Checkpoint loguniq 28453, logpos 0x2b96018 09:50:50 Checkpoint Completed: duration was 126 seconds. 09:50:50 Checkpoint loguniq 28453, logpos 0x2cdc018 09:57:03 Checkpoint Completed: duration was 73 seconds. 09:57:03 Checkpoint loguniq 28453, logpos 0x32c1018 10:04:05 Checkpoint Completed: duration was 122 seconds. 10:04:05 Checkpoint loguniq 28453, logpos 0x3df4018 How can i find the reason for the checkpoints and solve the problem? I know that some users are starting calculating jobs which will delete/write about 500000 rows within one transaction maximum. There are also using a lot of temp tables and do insert/updates within the temp tables. Greetings, J'rg sending to informix-list sending to informix-list