Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Eugene, So in your opinion (stress option) is this unrelated to DRBD and more of an EXT3/MD/Kernel issue? Were you using MD or hardware raid when it failed? What are you using now? Did you ever try a different filesystem other than EXT3 on top of MD/DRBD? Thanks, Brent At 03:56 PM 7/11/2006 +0400, Eugene Crosser wrote: >Lars Ellenberg wrote: > > / 2006-07-10 10:16:17 -0700 > > \ Brent Jensen: > >> The filesystem was created correctly on the drbd0 device and > >> verified that this is the case. This is the same issue as shown in: > >> http://lists.linbit.com/pipermail/drbd-user/2005-October/003860.html > > > > If it was the same issue, it would have been fixed by using some > > kernel without the bio-clone bug. We did not hear from Eugene > > about this since, and, iirc, he does thorough tests in a well > > established environment. > >We've just quit using the hardware that demonstrated that behavior. > >I was never able to pin down the problem to any specific element; I do >know that exactly same version combination of the kernel+drbd work fine >on some hardware and give in-memory filesystem corruption on other. > >In particular, Dell 1U servers with Megaraid SCSI RAID controller never >gave me any trouble. > > >> This allusive bug has never gone away (I've been experiencing > >> this for almost two years now). I would like to know how people > >> are dealing w/ this. It has recently become a real problem in my > >> production environment (almost every day). I created the > >> filesystem on both production setups on the DRBD device. They > >> might seem to fail (but not always) when heavily used (lots of > >> web activity or during backups). > > > > In each case I personally have seen fs-corruption occur like this, > > appart from that bio-clone bug, it finally could be tracked down > > to bad hardware. bad cable, cable interferrence, instability in > > voltage (some people could reproduce fs-corruption by running > > 3D-benchmarks on their grafic card)... > > > > There may be problem deeper in the kernel, there are some threads > > about fs corruption with devmapper, you could try to find > > something relevant for your situation on e.g. > > http://groups.google.com/group/fa.linux.kernel/ > >Eugene > > > >_______________________________________________ >drbd-user mailing list >drbd-user at lists.linbit.com >http://lists.linbit.com/mailman/listinfo/drbd-user