[566] in Athena Bugs
[Mark W. Eichin: adb and the mount table]
daemon@ATHENA.MIT.EDU (Ken Raeburn)
Thu Jul 28 03:13:32 1988
Date: Thu, 28 Jul 88 03:12:54 EDT
From: Ken Raeburn <raeburn@ATHENA.MIT.EDU>
To: bugs@ATHENA.MIT.EDU
In general, I think we need to update the adb macro set for our
current system. (The new NFS release does nothing about the adb
macros.) Mark's changes will be a good start. (He did this in trying
to get rid of a hard-mounted file system from LCS when a gateway went
south.)
------- Forwarded Message
Date: Wed, 27 Jul 88 06:05:55 EDT
From: Mark W. Eichin <eichin@ATHENA.MIT.EDU>
To: Ken Raeburn <raeburn@ATHENA.MIT.EDU>
Subject: adb and the mount table
Well, I got it to time out, though 'mount' still says it is
hard mounted... I now have adb macros for printing sockaddr_in, struct
vfs, and something *like* the mntinfo structure (but not quite...)
(look in eichin:adb) It should be a mntinfo structure, but the fields
just don't seem to match. Enough were there for me to extrapolate, and
forcing a soft mount of another directory was enough to determine what
bitfields to try changing. Sure enough, one of them got timeout errors
to happen, and I got back my emacs and an 'ls' that I had used as a
probe.
It seems that _rootvfs is always the real root filesystem, but
that everything that gets mounted after that gets inserted in slot 2,
between it and the rest of the linked list.
I just got it unmounted too... just had RPC_PMAP_FAILURE -
RPC_TIMED_OUT and then it was happy.
I still haven't got a case of the console problem to happen...
I will logout of kremvax now and see what it does.
_Mark_
------- End Forwarded Message
