Re: Arla 0.42-RC2

Tomas Olsson <[email protected]> writes:

> > (I configured arla with --with-pthreads). I do have a file
> > /lib/libgcc_s.so.1 installed, so this error message seems strange.
> > 
> That message is probably from the pthreads implementation, so I guess we
> have a linking problem. Things to try:

I'll look further into this tomorrow. What's confusing, to me, is that
I'd expect linking problems to show up earlier, at load-time. It seems
like some dynamic/runtime loading is happening here, either explicitly
by arla, or by some glibc internals.

> >   ls: /afs: No such device
> > 
> > Is there any more appropriate error code that could be returned?
> > ENODEV made me suspect the nnpfs module, which at the moment seems to
> > be innocent.
> >
> I don't know. The rationale would be that nnpfs depends on its daemon like
> an ordinary fs depends on its disk. Any ideas?

Not sure. If a disk is broken, I think you would get EIO, not ENODEV.
I've only seen ENODEV when you try to open a device node, and the
kernel doesn't recognize the device numbers. Some candidate errno
values: ENOTCONN, EPIPE, EIO.

Regards,
/Niels