I don't know why this has not happened before but I'm running a recent
Squeeze system and this race condition manifests itself when the checkfs
script fails, leaving /var/log/fsck/checkfs with cryptiv messages like:
"fsck from util-linux-ng 2.16.2" "fsck.ext3: no such file or directory"
My boot fails, and my USB disks are not mounted. Which kinda sucks
because /usr is a USB SSD.
It took me a while to figure out that "fsck" (2.16.2) was actually
finding "fsck.ext3", it was "fsck.ext3" that could not find my USB disk
because it had not been fully blessed by what ever glacial process we're
now using for this purpose.
I fixed the problem by putting a "sleep 20" into "checkfs.sh". It
sucks. But what else can I do? I'd rip out hald if I could but instead
of helping here, it's just another impediment.
It seems that there have been a boatload of swell updates that have made
upgrading to Squeeze exceptionally painful, and this is just another
one. Like evdev leaving my system with inert as a brick since X won't
take the keyboard and X won't let anyone else take it, so I have to
power cycle the machine, and hope my disks are recoverable. Great move
forward.
I notice from the lack of Debian responses in 7 months, no one seems to
care. Whatever. So much for "no bug too deep," instead it's "lets make
Firefox lamer by calling it Iceweasel - but the binary is still
firefox-bin" or "let's break cdrecord by replacing it with 'wodim' which
was abandoned years ago". Nice work.
--
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]
