Hi there, Aren't you able to turn write-bacl off (thus enabling write-though) by using hdparm? []s, Fernando Lozano > Frank Warnke wrote: > > > Could you elaborate on why "IDE disk/controllers and journal fs are not > > recommended for DB usage"? Does "DB" mean database? Does this blanket > > statement cover ATA RAID controllers as well? Is this a performance > > issue or a data loss issue? > > yes. > yes, DB = data base. > yes, it covers ATA RAID controllers. > it's a data loss issue. > > main problems: > ATA disks _manufacturers_ enable write-back-caching > and > ATA disks allow to reorder the operations in write-back-cache > > There a lot of literature about that: > http://sr5tech.com/write_back_cache_experiments.htm > http://sr5tech.com/download/ATA_RAID5_Whitepaper.pdf > http://archives.postgresql.org/pgsql-admin/2000-12/msg00094.php > http://redhat.com/archives/fedora-test-list/2003-November/msg01447.html > http://redhat.com/archives/fedora-test-list/2003-December/msg00001.html > http://marc.theaimsgroup.com/?t=104991701200010&r=1&w=2 > http://marc.theaimsgroup.com/?t=107014526000001&r=2&w=2 > http://marc.theaimsgroup.com/?l=linux-kernel&m=106745860511194&w=2 > > > I also have over 40 servers in the field using either Dell(LSI) or 3ware > > PATA/SATA hardware RAID controllers (RAID5) running databases on > > ReiserFS and EXT3 filesystems. Many have seen close to two years of > > service and they have been problem free. Our customers are very pleased > > with the performance of the database access. > > if you did not have any problem, it doesn't guarantee that it can not > happen > > > People, including myself, come here to learn and when we can, add our > > experiences. Please be careful with your opinions, there are a lot of > > people viewing this list because they assume you have the answers based > > on experience. It would help a viewer of this thread to see a reference > > or two to substantiate your claim. If I have done a customer a > > disservice in the configuration of their servers, I want to remedy it > > immediately. > > Those are no personal opinions, they are real facts. > My opinions are mine, if you don't like them you are free to delete the > e-mail and forget it :-) > > > -- > fedora-list mailing list > fedora-list@xxxxxxxxxx > To unsubscribe: http://www.redhat.com/mailman/listinfo/fedora-list