Il giorno gio, 01/12/2005 alle 21.33 +0100, Jan Engelhardt ha scritto:
> >> >> Is this a known issue and is there a solution or workaround by any
> >> >> chance?

> 
> Right. The first thing is, if expanded, a serious pivoting I could not 
> understand as long as I had it in the inbox. It was a privately sent one, 
> so there is no record in the archives.
> I think it was like this solution
>   http://www.fsl.cs.sunysb.edu/pipermail/unionfs/2005-October/001305.html
> the unexplainable part being the
>   pivot_root /ram/union /ram/union

i remember i met Jürg Billeter in the chan #unionfs, he is using
run_init as me, i don't know if this can be the reason, and if that
issue is not present using pivot_root and moving /proc, but for sure it
exists using our way to change root.

As initramfs and klibc run_init seem that will substitute initrd and
pivot_root i think it would be nice to figure out a solution for this.

why for ex /proc/1/exe is a link to /real/path/to/sbin/init
and /proc/1/root is a link to unionfs root [/]?

If this issue affects debian's start-stop-daemon, then it could be a
really big problem for the stability of a livecd debian based, unionfs
allows to install new packages, and some of them will probably fail
running postinst scripts.

_______________________________________________
unionfs mailing list
[email protected]
http://www.fsl.cs.sunysb.edu/mailman/listinfo/unionfs

Reply via email to