Re: checkpoint interval

What are your LRU_MAX_DIRTY and LRU_MIN_DIRTY set to? We reduced them to 1 and 0 respectively to keep the buffers as empty of dirty pages as possible. This speeds up checkpoints. Also take a look at your LRU queues and see if the cleaner threads are able to keep the queues as empty as they are supposed to (onstat -R). -- dbs -- David Samson, Senior Database Researcher dsamson@ndsisrael.com NDS Technologies Israel Ltd. +972 2-589-4529 PO Box 23012 Fax: +972 2-589-4578 Jerusalem, Israel Cellular: +972 51 287 481 Harold Luse <hluse@vth.colostate.edu> wrote in article <6hqfj6$jk9$1@news.xmission.com>... > > Sahrul, > > We have been having similar problems with 7.23 UC6 on AIX 4.2.1. > > Our routine checkpoint intervals are no more than 2 seconds. However, > approxiamtely two to three times each day, we get excessive checkpoint > intervals of anywhere from 80 to 250 seconds. > > Through the use of a shell script and some onstat commands we have > tracked the problem down to the logical buffer not being able to flush to > the logical logs on disk (we use unbuffered transaction logging). > > Are you noticing the same thing? > > Anyone else who cares to comment please jump in. > > ==================================================================== > Harold Luse Phone: (970) 491-4120 > Veterinary Teaching Hospital Fax: (970) 491-1275 > Colorado State University Pager: (970) 229-8173 > Fort Collins, Colorado USA E-mail: hluse@vth.colostate.edu > ==================================================================== > >