Re: [Dazuko-devel] Some bugs detected

[John Ogness]

Tikka, Sami wrote:
> It seems to me that the options --enable-chroot-support and
> --disable-chroot-support are reversed. If I want to enable chroot support, I
> have to use --disable-chroot-support :)

>From a kernel developer perspective, you may be correct (at least with how
the define is named). But from a user perspective, I think it is correct how
it is. The output of "./configure --help" states:

--disable-chroot-support      disable resolving chroot'd filenames

and that is (in my opinon) correct. The fact that it then leads to defining
"USE_CHROOT" could cause confusion for developers, but users don't see that.

By the way, I chose the define "USE_CHROOT" because it tells the Dazuko code
that it should activate the "chroot:" file namespace protocol (which should
only be activated when chroot'd filename shouldn't be resolved).


> We have also had a couple of kernel oopses with the latest dazuko but that
> might be a configuration problem (using local dpath on smp machine and
> such... I'm checking it out tomorrow.)

I assume you are using syscall hooking for 2.6. The OOPS's are probably
related to this (not enough stack space with execve() ?). It would be great
if you could reliably reproduce the problem. And then try disabling certain
events (such as ON_EXEC) to see if the problem goes away. (ON_EXEC has
caused problems in the past with syscall hooking on Linux.)


> Then there is a problem with return values from dup and dup2 calls. Possibly
> with close too. You should get a patch soon, tomorrow, I hope.

It would be nice if you could also provide scripts or test programs that
reproduce the problem.


> So, please hold off with 2.3 a couple more days.

Holding...

John Ogness

-- 
Dazuko Maintainer