Microvax 3100 Disk doesn't mount

[Renato Ramos]

Hi,

In our Organization we have a MicroVax3100, and one of disks doesn't mount.

Is there any way to try Up this drive and recover yours data ?

Thanks.

[abrsvc]

In order to properly provide assistance here, please post the commands used to try to mount the disk and the output lines the system generated from that attempt.
Also post the OpenVMS version of the system

Thanks,
Dan

[Bob Gezelter]

On Wednesday, July 29, 2015 at 11:28:48 AM UTC-4, Renato Ramos wrote:

Rentao,

It depends. Precisely what error message(s) are you getting? Is the drive spinning?

- Bob Gezelter, http://www.rlgsc.com

[Renato Ramos]

I did 3 reboots and the drive doesn't boot. (not mounted)

[abrsvc]

I this disk the system disk? What messages are seen on the console when attempting to boot?

[Renato Ramos]

On Wednesday, July 29, 2015 at 1:00:54 PM UTC-3, abrsvc wrote:
> I this disk the system disk? What messages are seen on the console when attempting to boot?

This device isn´t the System disk:


$ sh dev DQB300/full

Disk MANTR$DQB300:, device type Generic SCSI disk, is online, file-oriented
device, shareable, error logging is enabled.

Error count 1 Operations completed 2
Owner process "" Owner UIC [0,0]
Owner process ID 00000000 Dev Prot S:RWED,O:RWED,G:RWED,W:RWED
Reference count 0 Default buffer size 512


Can you recommend any command to try mount this device ?

[abrsvc]

Assuming that you are using the system account, try the command below and posst the results:

$ MOUNT/OVER=ID DQB300

[Renato Ramos]

On Wednesday, July 29, 2015 at 2:29:23 PM UTC-3, abrsvc wrote:
> Assuming that you are using the system account, try the command below and posst the results:
>
> $ MOUNT/OVER=ID DQB300

I got this message:

%MOUNT-I-OPRQST, Please mount device _MANTR$DQB300:

[VAXman-]

In article <095ef081-62c4-4122...@googlegroups.com>, Renato Ramos <contr...@gmail.com> writes:
>On Wednesday, July 29, 2015 at 1:00:54 PM UTC-3, abrsvc wrote:

>> I this disk the system disk? What messages are seen on the console when =
>attempting to boot?
>
>
>This device isn=B4t the System disk:

>
>
>$ sh dev DQB300/full
>
>Disk MANTR$DQB300:, device type Generic SCSI disk, is online, file-oriented
> device, shareable, error logging is enabled.
>

> Error count 1 Operations completed =
> 2
> Owner process "" Owner UIC [=
>0,0]
> Owner process ID 00000000 Dev Prot S:RWED,O:RWED,G:RWED,W:=
>RWED
> Reference count 0 Default buffer size =

> 512
>
>
>Can you recommend any command to try mount this device ?

$ MOUNT/OVERRIDE=IDENTIFICATION MANTR$DQB300

--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.

[abrsvc]

The above message suggests to me that this is a hardware issue that may not be easily resolved remotely (ie. via messages here).

There are essentially 2 components to the hard drive: the drive itself and the controller which resides on the drive (the board attached to it). I believe that the controller has a problem. VMS recognizes the presence of the controller and its ID which is clear since you can "see" the device. Any attempt to access it however, seem to indicate that it is not there, thus the "please mount" message.

Where are you located? This may be a problem with the drive controller or the motherboard of the VAX itself.

Dan

[Bob Gezelter]

On Wednesday, July 29, 2015 at 1:34:11 PM UTC-4, Renato Ramos wrote:

Renato,

If you are getting the "Please mount device" message, the drive is likely not operating properly.

If the file system was merely damaged, MOUNT/OVERRIDE=ID would generally work. You can verify by attempting to do a MOUNT/FOREIGN on the device, but I would expect it to similarly fail.

The data might be intact, but that would require some careful work in a clean room repair facility. Do you have recent backups? There are a variety of options here, but care is required. It is very easy to make the situation far worse.

If you wish to speak offline, please let me know via email.

[Scott Dorsey]

Renato Ramos <contr...@gmail.com> wrote:
>I did 3 reboots and the drive doesn't boot. (not mounted)

That's nice. So what is this drive and what is this operating system?
Does the drive show up when you do a "SHOW DEV D" from the monitor prompt?
Does the drive spin up? Has it been eaten by walruses?
Do you have a good backup?
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

Renato Ramos <contr...@gmail.com> wrote:
>On Wednesday, July 29, 2015 at 1:00:54 PM UTC-3, abrsvc wrote:

>> I this disk the system disk? What messages are seen on the console when =
>attempting to boot?
>
>
>This device isn=B4t the System disk:

>
>
>$ sh dev DQB300/full
>
>Disk MANTR$DQB300:, device type Generic SCSI disk, is online, file-oriented
> device, shareable, error logging is enabled.
>

> Error count 1 Operations completed =
> 2
> Owner process "" Owner UIC [=
>0,0]
> Owner process ID 00000000 Dev Prot S:RWED,O:RWED,G:RWED,W:=
>RWED
> Reference count 0 Default buffer size =

> 512
>
>
>Can you recommend any command to try mount this device ?

You mean like "MOUNT DQB300:" for instance?

[Scott Dorsey]

On Wednesday, July 29, 2015 at 1:34:11 PM UTC-4, Renato Ramos wrote:

Assuming the SCSI buss is okay and the termination isn't screwed up somewhere
(which are the first things to check), the drive is shot. Replace it,
reformat, and restore from backup.

[Michael Moroney]

Try the following, to see what the real error is.

$ MOUNT/OVER=ID/NOASSIST DQB300

Probably DEVOFFLINE, and the drive is probably shot.

[Bob Gezelter]

Scott makes a good point. Bad cables/termination can present as a non-functioning drive. Replacing old cables and terminators may be a worthwhile exercise.

[John Reagan]

Not sure that replacement is needed, but if somebody has bumped the machine around (cleaning staff, IT folks running a nearby cable to some other systems, etc.) then you can easily dislodge a cable/terminator. I would certainly suggest reseating all the cables.

[abrsvc]

I have successfully "recovered" drives like this by swapping the controller cards with a known good drive of the same model. I only run this way long enough to recover the data though with mounting the drive as read-only.

Dan

[Renato Ramos]

On Wednesday, July 29, 2015 at 4:35:20 PM UTC-3, abrsvc wrote:
> I have successfully "recovered" drives like this by swapping the controller cards with a known good drive of the same model. I only run this way long enough to recover the data though with mounting the drive as read-only.
>
> Dan

Thanks a lot for all recommendations.
I´ll try open the machine and clean/verify all cables.

[terry+go...@tmk.com]

On Wednesday, July 29, 2015 at 3:53:05 PM UTC-4, Renato Ramos wrote:
> Thanks a lot for all recommendations.
> I´ll try open the machine and clean/verify all cables.

Since this is a VAX*, you could try $ANALYZE/ERROR/SINCE=YESTERDAY/OUT=somefile.txt and then see if there are any useful messages in there. You'll have timestamps and other unrelated info. I deliberately didn't specify including only the drive device name as the error could be on the PKxx controller device as well. I said YESTERDAY on the assumption that you tried mounting it more recently. If not, specify some date in the past, before you ran into problems. The output file will be correspondingly larger.

Once you've looked at the log and verified it doesn't contain any data you want public, upload the text logfile to someplace like pastebin and post a link to it here.

* If it was an Alpha or Itanium, you'd be suffering through the DIAGNOSE command, but an old MV3100 should be perfectly happy doing an ANALYZE.

[Chris Scheers]

OK, I have not played with all the MicroVAX 3100 models, so I may be
missing something, but shouldn't a MicroVAX 3100 SCSI drive be a DK drive?

Isn't a DQ drive an IDE drive?

What model MicroVAX 3100 is this? How is the drive connected?

--
-----------------------------------------------------------------------
Chris Scheers, Applied Synergy, Inc.

Voice: 817-237-3360 Internet: ch...@applied-synergy.com
Fax: 817-237-3074

[Hans Vlems]

Yes, even though the show device dqa300 command output lists it as a generic SCSI device.
Perhaps this is one of those 3100's that also support DSSI disks (and become a VAX 4100). IIRC the main boards support connectors for DSSI and SCSI. They only differ in color. If so the cable may have been put in the wrong connector.
DQ devices use a single digit, 0, or 1, So dqb300 is strange for the 3 and the triple digits.

[Bob Koehler]

Yep.

You need a more detailed answer, you'll need to provide a more
detailed description.

[Bob Koehler]

In article <5529281b-723f-470e...@googlegroups.com>, Renato Ramos <contr...@gmail.com> writes:
> On Wednesday, July 29, 2015 at 4:35:20 PM UTC-3, abrsvc wrote:

>> I have successfully "recovered" drives like this by swapping the controll=
> er cards with a known good drive of the same model. I only run this way lo=

> ng enough to recover the data though with mounting the drive as read-only.

>>=20

>> Dan
>
> Thanks a lot for all recommendations.

> I=B4ll try open the machine and clean/verify all cables.

Generally when drives fail, it's the electronics, not the storage.
There a commercial folks working in this area which will temporarily
hook up working electroncs, pull the data for you, and ship you the data.

[johnwa...@yahoo.co.uk]

Thank you, well spotted (unless matters have changed since I last
played with real 3100s).

Given that clear answers to apparently simple questions like "how
is this device connected" or "has this command ever worked?" can
sometimes be surprisingly difficult, are there any simple ">>> mode"
or VMS commands whose output would be helpful here?

"$ SHOW DEV D" springs to mind for disks, but maybe there are better
options.

It's been a long day and brain fade/lack of time prevents me posting
the DCL to show definitively what kind of box VMS thinks it is; all
I know is that it's easy :( Suggestions?

A working DQ on a 3100 with no third party hardware would surely
be a surprise (?), so "how is this device connected" and ideally
"has this command/this drive ever worked" are still worth having
answered.

Fingers crossed.

[David Froble]

johnwa...@yahoo.co.uk wrote:

> It's been a long day and brain fade/lack of time prevents me posting
> the DCL to show definitively what kind of box VMS thinks it is; all
> I know is that it's easy :( Suggestions?

This is one method:

$ show license /charge
VMS/LMF Charge Information for node DFE90A
This is a VAXstation 4000-90A, hardware model type 475

[JF Mezei]

On 15-07-29 15:53, Renato Ramos wrote:

> Thanks a lot for all recommendations.
> I´ll try open the machine and clean/verify all cables.

From >>> does SHOW DEV show the drive ?

If so, then the communication between the SCSI driver and the SCSI part
of the drive works.


Once booted, you can try:

MOUNT/FOREIGN <disk>
This would check communication with the drive, but not the format of the
data on it.

BACKUP/PHYSICAL disk: otherdisk:[dir]temp.back/save

The above will make copies of each block in the "disk:" irrespective of
file format. You will see if there are errors reading individual blocks.

(you can specify NLA0:/SAVE as output if you just want to check if the
disk can be read.)

Also, when having disk problems, recording how much time it takes for an
error message to be issued matters. A very fast error message is likely
due to communications or failure of controller, whereas a long time
could be the disk trying and trying and tryiong before reporting an error.

[Hans Vlems]

Since we haven't heard from the OP for some time my guess is that most of his problems are owing to a typo: dQb300 vs dKb300...