From: jonathan@rubis.geop.ubc.ca (Jonathan Thornburg) Subject: extended file system + efsck + /dev/*/* (was: Re: [QUERY] Is efsck obsolete?) Date: 27 Oct 1992 16:17:40 GMT
In article <2AED36FD.46F5@tct.com> chip@tct.com (Chip Salzenberg) writes:
>First, the good news: SLS installation worked right the first time!
>Next, the better news: I have lots of free disk space!!
>
>However, there is trouble in paradise. I'm using the extfs (because
>fourteen-character limits are for playwrights), and everything seems
>to work fine... except that efsck has a conniption fit when I run it
>against my root partition, even when I boot from the SLS A1 disk.
>
>So is SLS's efsck out of sync with the actual behavior of the extfs?
>Or is my root filesystem really out to lunch? Perspiring minds want
>to know!
I installed the MCC 0.97pl2 release about 3 weeks ago. Everything
went pretty smoothly, and it has worked perfectly from day 0. I
too opted to use the extended file system for everything (I agree
with Chip on the merits of 14- (and 8- :-( ) char limits), including
my working root partition. (I also have an inactive minix-fs root
partition hidden away as disaster insurance.)
I have had the same efsck problem Chip reports -- efsck goes berserk
when run against my root file system. Check the inodes it's complaining
about with 'find / -inum 1234 -print' etc, they all turn out to be
device spcial files.
Conclusion:
(1) IMHO, there's some bug in efsck's handling of device inodes
which causes efsck to cry wolf on them.
(2) I haven't tried it, but it would be pretty trivial to egrep -v
the efsck output to filter out the /dev/*/* inodes.
- Jonathan Thornburg
<jonathan@geop.ubc.ca> through mid-November 92, then
<jonathan@einstein.ph.utexas.edu> or <jonathan@hermes.chpc.utexas.edu>
[for a few more months] UBC / {Astronomy,Physics}
[then through Aug/92] U of Texas at Austin / Physics Dept
/ Center for Relativity