On Mon 2002-05-13 (14:09), Doug White wrote: > On 13 May 2002, David [ISO-8859-1] Siebörger wrote: > > > That's hardly the worst of it. The ServerWorks OSB4 ATA controller > > has been known to cause data corruption with Seagate drives. > > Have you isolated it to Seagates only? our problematic rackables have > seagate drives, but I don't have any other mfr drives (maxtor, ibm) to > test it with.
No, that's from the Alan Cox posting I mentioned. I'm afraid didn't have any drives of other brands handy so I gave up on the IDE interface and went all-SCSI. Here's another report of the problem, but also with a Seagate disk: http://docs.freebsd.org/cgi/getmsg.cgi?fetch=577071+0+archive/2002/freebsd-hackers/20020224.freebsd-hackers > > I've just had a very frustrating experience with this problem on an > > Intel STL2-based 4.6-PRERELEASE system which experienced segfaults > > and file system corruption. We tested the memory, checked the PSU, > > heat in the system, replaced the IDE cable, replaced the drive > > (unfortunately with another Seagate!) before I heard about this > > issue. My testing showed that the problem only occurred while > > writing to disk using UltraDMA. > > On the SCB2 Intel has a big warning in the manual to not use that IDE > interface for system disks. I don't see it in the STL2 manual so I guess > they learned their lesson later on. We don't have any SCB2 systems here, but its specs say it has a Promise controller, rather than the ServerWorks chip on our STL2. -- David Siebörger [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message