Page Cleaners Timing out

> > > > we are running online ver 5.00.UC2 with i-star and are experiencing a > > problem with the page cleaners timing out. we are running on a sun > > 690 with the database accelerator kernal enhancement. > > > > a tbstat -u gives the following: > > > > (hermes1)/home_hermes1/baskett>tbstat -u > > > > does anyone know what is causing this? seen this before? any fixes? > > i havent posed this question to tech support yet. Bob, Yes, we have seen this behavior under 4.0 and 4.1 Online. I believe this happens when the page cleaners cannot get their work completed in a specified amount of time. I do not know how to tweak the time a page cleaner has to do its work. Anyone know? Under 4.1 and 5.0, you can adjust LRU_MIN_DIRTY and LRU_MAX_DIRTY in your tbconfig so that the page cleaners have somewhat less work to do. The default values for MIN and MAX are 50% and 60%, respectively. I'd suggest lowering them to fire off when around 20% of the pages are dirty, but you need to experiment some, as your situation may be different than ours. I would also increase the checkpoint interval at the same time. The optimum setup is for the page cleaners to fire off and complete their work, just prior to the checkpoint. Since shared memory cannot be changed while a checkpoint is in progress, i'd say it is much more beneficial for the flushing work not to be initiated by the checkpoint. This scenario can be monitored by running 'tbstat -R' and 'tbstat -F'. The 5.0 Online Administrators manual has an excellent writeup on this subject. Please let me know what you discover. I dont believe I know all their is to know about this subject, but i'm learning.... Naomi -- Naomi Walker (aka N7FSA) Anasazi Inc. Phoenix, Arizona Enthusiasm is caught, In tennis, not taught. love means nothing.