I manage a DEC-400-4/233, running OSF/1 v3.2 (upgraded from 3.0B).
Since the upgrade I get this message when making a AdvFS filedomain:
# mkfdmn /dev/rz13e econprog_domain
ioctl (WDINFO): No space left on device
I, by some reason, do not get the message if the domain is made on the
'a'-partition (take that with some salt).
The domain is created and I am able to make filesets on it, everything
seems to work all right. Somone out there having this problem?
I looked around a bit, and the disklabel(4) manpage states this:
[...]
Finally, the DIOCWDINFO ioctl operation sets the in-core label and then
updates the on-disk label; there must be an existing label on the disk for
this operation to succeed. Thus, the initial label for a disk or disk pack
must be installed by writing to the raw disk. All of these operations are
normally done using the disklabel program.
Which may indicate that the disklabel-program have not done its job
properly, does it have to?
What I usually do when installing a disk is:
I allways have all SCSI-devices defined in the kernel config-file.
When a new disk arrives, I chains it in to the SCSI-chain on the fly.
Then I disklabel it (eg. rz13):
disklabel -wr -t advfs /dev/rrz13c foo
This does not produce any error messages. Then I partition the disk,
making sure no overlap.
Then I do:
# mkfdmn /dev/rz13e econprog_domain
and have started to get:
ioctl (WDINFO): No space left on device
# mkfset econprog_domain econprog
# mkdir /mountpoint
mount econprog_domain#econprog /mountpoint
Am I to optimitic about something here? I do not think so.
Giving
# disklabel /dev/rrz13c
and
# disklabel -r /dev/rrz13c
gives the same results, which should indicate that I have succeeded in
writing the lable to disk?
None of our other Alphas produce this error message. Any ideas? (It
can not just be that 13 is my unluky number, can it ;-)
Morten
Received on Fri Apr 21 1995 - 12:07:25 NZST