RE: Extremely high uptime caused by oninit processes

Time to start digging into the ONSTAT command :-) You need to find out what those oninit's are doing - or in other words, what Informix is doing. I usually start by seeing if anyone is running a big (batch) process that could slow the system down a lot (mine is a TP system). Are the top processes on the machine Informix only (oninit) ? -----Original Message----- From: owner-informix-list@iiug.org [mailto:owner-informix-list@iiug.org] On Behalf Of julio@prodesmaq.com.br Sent: 07 March 2005 02:21 PM To: informix-list@iiug.org Subject: Extremely high uptime caused by oninit processes Hi guys, Since last week, I'm experiencing a very big problem with my db server performance. The uptime output shows me values even higher than 25! I had never seen such thing before. To complicate even more, sometimes when a look at the uptime it's under 2, sometimes under 1 and suddenly this value jumps to 10, 15 or some higher value. As I'm kinda lost searching for the problem, I would be very pleased if someone could at least point me to the right direction to find the problem. I'm using: Suse Linux 9.1 Professional Kernel 2.6.5-7.104-smp IDS 7.31UD1 I run update statistics every day, dropping the distributions and executing a medium update for all tables and a low update in all indexes heads. Some time ago I had the same sintoms but it was a problem with the ER replication (related to the size of my dict pool). Since I do not use the replication anymore, this should not be the problem, should it? Should it be some problem with some table or some index? Should I send some command output, like a onstat -a or something? Any help would be very, very appreciated. Thanks all. Julio Cesar sending to informix-list sending to informix-list