Re: [NetBSD 8.0] Another panic in ffs_newvnode()

[BERTRAND Joël <joel.bertrand%systella.fr@localhost>]

Michael van Elst a écrit :
> joel.bertrand%systella.fr@localhost (=?UTF-8?Q?BERTRAND_Jo=c3=abl?=) writes:
>
> > Michael van Elst a écrit :
> > > joel.bertrand%systella.fr@localhost (=?UTF-8?Q?BERTRAND_Jo=c3=abl?=) writes:
> > >
> > > > 	Full output is very large...
> > >
> > > Check the output for cg 0, in particular wether inodes 0-2 are
> > > in use.
>
> > 	I'm not a specialist of fs issues. How can I do that ?
>
>
> That's the in the "very large" full output. For example:
>
> cg 0:
> magic   90255   tell    6000    time    Thu Jan  1 01:00:00 1970
> cgx     0       niblk   0       ndblk   8192
> nbfree  956     ndir    0       nifree  1918    nffree  7
> rotor   0       irotor  0       frotor  0
> frsum   0       0       0       0       0       0       1
> sum of frsum: 7
> clusters 1-8:   0       0       0       0       0       0       0       0
> clusters 9-15:  0       0       0       0       0       0       0
> clusters size 16 and over: 1
> clusters free:  68-1023
> iused:  0-1
> free:   537-8191
>
> Shows that inodes 0-1 are allocated ("iused"). However, FFS must at least have
> inodes 0-2 allocated (2 is the root directory).
>
> N.B. fsck _does_ correct this, but of course not if it assumes the FS is
> clean, use 'fsck -f' to force a real check.

	I have checked this fs with fsck -f before sending my first mail. I 
have fixed a fts_read error (I have seen a lot of fts errors with early 
7.0) with clri and fsck. I hope it will be enough...

	Regards,

	JKB