This is a discussion on Re: Long fuzzy checkpoint within the Informix forums, part of the Database Server Software category; --> I think is unlikely that the problem is related to the fact this is a "fuzzy" checkpoint. If you ...
| |||||||
| Register | FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
| ||||
| I think is unlikely that the problem is related to the fact this is a "fuzzy" checkpoint. If you don't see that the prolem is I/O, maybe th eproblem is that a thread is in critical section (look in onstat -u for a thread with the "X" flag). Then try to analyze what is this thread doing. Gustavo Omar Muņoz wrote: > Hi. > > I've seen on my site that sometimes a fuzzy > checkpoint takes more than a minute to end. We have > configured LRU limits in order to activate cleaners (2 > and 1% for 384 LRUs. There are 950000 buffers of 2 Kb > each) and 5 minutes of checkpoint interval. Problem > persists, and it doesn't seem to be a writing item. In > fact, there was a 75 sec checkpoint for only about > 4000 dirty buffers (on onstat -R), when sometimes > there are more than 9000 and checkpoint doesn't take > more than 10 secs on those cases. I suppose physical > log doesn't have nothing to do also, because the > problem is fuzzy checkpoint. I ask you for your advice > about which other factors may be provoking that > situation, so I can make up configuration. > > Everything is about a Solaris 9 Server (8 > processors) and IDS 9.40 FC7 > > Thanks in advace > > Omar Muņoz > > > > __________________________________________________ __________________________________ > Looking for last minute shopping deals? > Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsea...egory=shopping > _______________________________________________ > Informix-list mailing list > Informix-list@iiug.org > http://www.iiug.org/mailman/listinfo/informix-list > > > |
| Thread Tools | |
| Display Modes | |
|
|