Re: Latest ext2fs Failure
Roland McGrath <email@example.com> wrote:
> Whenever you see this error "Translator died", then what you should do
> is manually set the active translator:
> settrans -a /jtobey /hurd/ext2fs /dev/hd0s6
> Then you should be able to see any messages from ext2fs come to the
> terminal where you ran settrans. It may be something straightforward
> causing ext2fs to fail to start up at all. If it does start up and then it
> crashes, then start it up, run gdb /hurd/ext2fs and attach to the right
> pid, suspend gdb, do "sh -c 'cd /jtobey' &" or whatever causes it to crash
> (putting the command in the background since it will block talking to the
> ext2fs process that you have sitting stopped in gdb--or you can do it on a
> different terminal). Then go back to gdb and do "cont"--you should see it
> stop where ext2fs is crashing; do "bt" there, and show us everything gdb said.
OK, but I think I'll have to reboot before I can get that far.
Currently, typing `reboot' gives
bash: ../sysdeps/mach/hurd/fork.c:158: __fork: Unexpected error: (ipc/send) invalid destination port.
and the terminal locks up (Bash crashing I suppose). On another
terminal I type `ps A' with the same results (except "-bash" instead
I also got this at some point:
ext2fs: ../../libdiskfs/disk-pager.c:93: fault_handler: Assertion `env && "unexpected fault on disk image"' failed.
I suspect that I should not have allowed Linux to fsck the Hurd root
partition. I'll reinstall. BTW, after running settrans --goaway
/jtobey and using settrans -a as you suggested, no output appeared in
the xterm that did settrans, just the "Translator died" message on the
one that did cd.
Well, now I've completely hung the system, so the reset button is