[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: another seriously corrupt ext3 -- pesky journal



In message <20030818181354 GC10270 think>, "Theodore Ts'o" writes:
> On Mon, Aug 18, 2003 at 12:39:46PM -0400, Erez Zadok wrote:
[...]
> What probably happened is that the power failed while you were writing
> out the inode table, and the memory failed before the DMA engine and
> hard drive did, since DRAM tends to be more sensitive to voltage drops
> that other parts of the system.  As a result, random garbage got
> scribbled all over the disk.  (Ted 's observation: PC Class hardware
> is sh*t.)

No kidding.  I have raid5, plus a spare, plus a UPS, and I still got hit.
People in the area said that there were such terrible power spikes that even
an average UPS might not have been able to totally filter those out.

Plus, my machine has 4GB RAM, meaning that even _more_ stuff was in memory
when things failed.  Sigh.

And while we're ranting, another personal lesson is that el-cheapo IDE raid5
systems should use *more* than one redundancy level (i.e., hotspares, raid6,
etc.).

> You can use debugfs's feature command to turn off the has_journal bit
> as follows:
> 
> debugfs -w /dev/hdaXX
> debugfs: features ^has_journal
> debugfs: features ^needs_recovery 
> debugfs: quit

I'll give that a try on a copy of the dd image file and let you know.  In
some of my experimentation this weekend, on another copy, I do recall that
just trying to get debugfs running gave me an error, and trying to open it
read/write failed.  So I may after all need your hacked version of debugfs.

Thanks,
Erez.




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]