Re: 5.0 OnLine - Process Aborted Abnormally - On New System

Posted in 1999

Robert Taylor — Wed, 28 Jul 1999 11:32:38 +0100 — source: Informix-list mailing list archive (1991-1998)

I take it there isn't any stupid developers / DBAs killing a process waiting on a latch or killing an SQL turbo ?

↪ replying to Robert Taylor

Mike Gee — Wed, 28 Jul 1999 09:14:11 -0400 — source: Informix-list mailing list archive (1991-1998)

Robert Taylor <robertt@scotlegal.com> wrote in message news:7nmmcn$d3t$8@plutonium.btinternet.com... > I take it there isn't any stupid developers / DBAs killing a process waiting > on a latch or killing an SQL turbo ? > > No, we only have 3 other guys on the system with ability to do that, and they all sit next to me, so I'm sure that's not the case. I've even checked the cron and various shell scripts for process kills and none of those seem to be the problem. One thing we did do was run a tbstat -a and pipe the output to a file every second, crossloaded the data in excel and graphed some of the data. We did notice that just before the DB crashed, it appears that we're getting a spike in the number of user_nreads and user_nwrites - but I don't know if that's directly related or not - could simply be a symptom of an sqlturbo w/ latch getting terminated for whatever reason.