> Which problem do you wrote as "unable to reproduce"?
Please simply ignore this, I was talking about two different things at
once. The problem can be reproduced only during LiveCD startup, using
busybox command with 'aufs' parameter.
> But aufs should work correctly which includes the expected error.
> Do you think aufs has no problem anymore?
> I am confused a little...
I'm sorry for confusing you. I tried so many variants that I mixed some
results before.
The final result is here:
Assume I add many branches to union, one by one, and then I call:
./busybox mount -o remount,ro aufs /union
(notice the 'aufs' parameter). This way, everything is OK in all cases,
- It doesn't matter if I have mount.aufs or not
- it doesn't matter if aufs is patched with the E2BIG patch
The command still works. I believe it's because this variant doesn't
call mount(2) with any additional arguments, only with NULL as a pointer
to 'const void *data'
But if I call
./busybox mount -o remount,ro /union
(notice the absence of 'aufs' argument in front of /union) then it
causes problem in LiveCD. The mount segfaults.
- It doesn't matter if I have mount.aufs or not
- it still doesn't matter if aufs is patched with the E2BIG patch.
The E2BIG patch has no effect regarding the segfaults in Live CD.
I have no explanation why it doesn't segfault in Installed Linux, even
if I use the same busybox binary and the same shared libraries.
Nevertheless, I believe it will be fixed if /proc/mounts doesn't contain
mount arguments, when you implement it through /sys.
The max amount of branches is still a big issue for me, I will be happy
if the max amount could be bigger than 127.
Tomas M
slax.org
-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
