On Sat, Dec 24, 2011 at 09:09:18AM +0000, Jamie Heilman wrote: > While I suppose technically this could be construed to be a "critical" > bug ... I'm not gonna call it that, because the autofs package is > pretty much a dead man walking at this point anyway (hell, it didn't > even make it into Squeeze, but it's still in unstable).
It's source appears to be in unstable, but not the binaries. I'll put a versioned Breaks on e.g. << 5 to allow for future renaming of autofs >= 5 to autofs. > So, in > relation to bug 494001 (the switch of /etc/mtab to a symlink to > /proc/mounts) this configuration makes autofs (*not* autofs5 mind > you... that works fine) fail in a rather obnoxious manner. This is > actually a known issue with autofs, although people are always > rediscovering it the hard way, and I imagine a few more will find out > shortly if they haven't already. See bugs 464436, 399614, 520095, and > for added flavor 588675 (which may not look related, but it kinda is > because /etc/mtab and /proc/mounts look really different when it comes > the root volume, and I didn't see that discussion brought up in > 494001, but it may bear thinking about as it's a potential source of > excitement). Anyway, my suggestion is to just add autofs to > initscript's Breaks: list and call it a day. Looks fairly abandoned, so I guess this is about the most realistic solution TBH. I'll add the Breaks. Regards, Roger -- .''`. Roger Leigh : :' : Debian GNU/Linux http://people.debian.org/~rleigh/ `. `' Printing on GNU/Linux? http://gutenprint.sourceforge.net/ `- GPG Public Key: 0x25BFB848 Please GPG sign your mail. _______________________________________________ Pkg-sysvinit-devel mailing list [email protected] http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/pkg-sysvinit-devel
