Subject: Re: (absence of) duplicated inodes
To: None <Christos-Zoulas@deshaw.com>
From: Duncan McEwan <duncan@Comp.VUW.AC.NZ>
List: current-users
Date: 09/23/1994 17:54:56
In message <199409220656.SAA08858@bats.comp.vuw.ac.nz> I wrote:

> Unfortunately I can't report the same good results.  I edited vfs_cluster.c
> to turn off doreallocblks, built (using sources supped yesterday) and booted
> a new kernel and then ran a perl script I wrote that previously relatively
> reliably triggered FS corruption.  It still does... :-(

I received the following response from Christos (I hope you don't mind me
posting this to the list Christos...)

> Try turning off doasyncfree too. It helped me :-)
> I have both turned off now.

That didn't work for me either.  My script still causes FS corruption after a
short running time... :-( Given that I seem able to cause the fault relatively
easily, can anyone suggest a good place to put some debugging to try to track
the problem down?

It occured to me after failing with both the doreallocblks and doasyncfree
suggestions that I have been having this problem since long before the
integration of the new 4.4 file system.  So if these variables that people are
tweaking were introduced at that time, they are not likely to help me.  Also
it seems odd that I should still have the same problem after such a major
change to the file system code, which is why (in an earlier message on this
subject, I mentioned that I had suspected my hardware...)

Duncan