Hello linux-kernel,
I'm using software raid5 on about 30 servers, and Yet twice I had a
serious data loss becouse of the behavior of linux RAID device.
In several cases I've got more then one of drives completely
disconnected. I have no ideas why this happened but this had
something to do with new AHA controler drivers and enabled TCQ.
After reboot all drivers turned back.
Other options for this to happen may be a controler failure, then
array is on several of them.
The linux RAID behavior in this case was to still allow write to the
array, so what only writes to online drives there made. This have
produced a huge mess on the drive after half an hour of running in
this mode.
The better solution I think would be block writes to the array
after second drive fail. This would at least give users more
recovery options.
-- Best regards, Peter mailto:pz@spylog.ru- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed May 23 2001 - 21:00:32 EST