> > -----Original Message----- > > From: linux-raid-owner@vger.kernel.org > > [mailto:linux-raid-owner@vger.kernel.org]On Behalf Of Neil Brown > > Sent: 17 November 2003 03:19 > > To: James R Bamford > > Cc: david.anderson@calixo.net; linux-raid@vger.kernel.org > > Subject: RE: Software Raid 1 data corruptions.. > > > > > > On Saturday November 15, jim@jimtreats.com wrote: > > <snip> Just so you don't feel bad, I have the following situation that is very similar -- used your test procedure to find the problem. Three systems with supermicro mbrds and DPT RA4050 controllers (raid not used). An assortment of Segate and IBM 18 gig drives. Setup is raid 1 with small boot & swap partitions + one large data partition. Symptoms -- raid 1 would build, but data corruption would occur either during initial data transfer or upon doing the first disk thrashing i.e. build a kernel or something like that. broke the raid down and tested by simply transfering a 30 meg file from an installed IDE drive to the scsi drive, dismounting, remounting and doing md5sum of the file. Consistently, the Segate drives come up with differing cksums. The IBM drives all seem to work as does the IDE drive. I have 9 Segate drives and they work in other machines so I have to assume it is some combination of the power supply, mbrd, controller, etc... that is causing the problem though I have 3 of these systems and it seems pretty consistent on all of them. Big mystery, you are not alone -- it may be your hardware. Michael Michael@Insulin-Pumpers.org - To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
