Highlighted
Absent Member.
Absent Member.
1052 views

NSS and/or disk Event Logging

Hi,

Is there any way to log low-level scsi disk events and NSS events? I am
trying to troubleshoot NSS in a NetWare 6.5sp8 guest on a vmware esx 3.5u4
host. So far I have not turned up anything in google or novell docs.

Thanks

Ron

Labels (1)
0 Likes
5 Replies
Highlighted
Knowledge Partner
Knowledge Partner

Re: NSS and/or disk Event Logging

Ron Neilly,
> Is there any way to log low-level scsi disk events and NSS events? I am
> trying to troubleshoot NSS in a NetWare 6.5sp8 guest on a vmware esx 3.5u4
> host. So far I have not turned up anything in google or novell docs.
>

What type of problems?

- Anders Gustafsson (Sysop)
The Aaland Islands (N60 E20)


Novell has a new enhancement request system,
or what is now known as the requirement portal.
If customers would like to give input in the upcoming
releases of Novell products then they should go to
http://www.novell.com/rms

0 Likes
Highlighted
Absent Member.
Absent Member.

Re: NSS and/or disk Event Logging

On Thu, 16 Jul 2009 20:47:17 +0000, Ron Neilly wrote:

> Hi,
>
> Is there any way to log low-level scsi disk events and NSS events? I am
> trying to troubleshoot NSS in a NetWare 6.5sp8 guest on a vmware esx
> 3.5u4 host. So far I have not turned up anything in google or novell
> docs.



You can load the scsihd.ham driver with a '/debug' parameter if you absolutely want to do so.
That would only be beneficial however if you can also interpret the output that this generates.

Outside a NSS Poolscan to have NSS development analyze that, in NSS there is not a lot of additional debugging you
can do.

Regards
Hans
0 Likes
Highlighted
Absent Member.
Absent Member.

Re: NSS and/or disk Event Logging

Well, obviously the type that need better reporting than that provided by
nss 😉

Alright, tongue-in-cheek sarcastic humour aside, I have a NetWare 6.5sp8
server in an ESX 3.5u4 vm. Twice in 7 days (same day of the week about the
same time of day during the processing of a trustee.nlm cronified job) NSS
reported a 'bad media' error. I have posted that issue in another thread a
few days ago on the 14th. So I won't repeat that here.

What I can't tell from the NSS event reporting is whether the problem
occured during a write (if so, then I would expect the hotfix feature of nss
to work around the bad block) or during a read. Was it during the writing of
a new block or the re-writing of an existing block? Was it a meta-data write
or a file write? Was is a write due to a salvage operation - ie a delete
that moves the data into a salvage tree? Etc.

Basically the NSS errortrapping and reporting is only marginally useful. I'd
also really like to know what events preceded the 'bad media' error. Did it
have a scsci communications error? Is there a problem with the driver? Etc.

I've come to the conclusion that NSS currently is not that capable, and
neither is NetWare. Probably if this were OES-Linux I'd have better event
trapping at the lower driver levels.

On the other hand Hans did let me know of the /debug option for scsihd.cdm
and I've put that in place. If any errors occur I may get some useful info
that I can pass on to our storage vendor (CoRAID) and/or Novell and/or
VMware.

Cheers,

Ron

0 Likes
Highlighted
Absent Member.
Absent Member.

Re: NSS and/or disk Event Logging

Thanks for the tip on the /debug parameter. I may or may not be able to
interpret the output. But even if I can't it may prove useful to others that
can interpret it, such as our storage vendor (CoRAID) or VMware (since we
are moving from physical to virtual on ESX 3.5u4) or Novell or LSI (driver
supplier).

There are many cooks in this kitchen and one of them just might be able to
figure out what the secret spice is. Eh?

Cheers,

Ron

0 Likes
Highlighted
Knowledge Partner
Knowledge Partner

Re: NSS and/or disk Event Logging

Hi,

Ron Neilly wrote:
>
> What I can't tell from the NSS event reporting is whether the problem
> occured during a write (if so, then I would expect the hotfix feature of nss
> to work around the bad block)


There is no such thing as a hotfix feature in NSS. Since minimum 10
years, hard disks do not produce OS-Aware write errors until they are
terminally broken and need to be replaced immediately. All block
reassigement since that time occurs in the firmware of the harddrive
itself (and can be read out through SMART). That said, block relocation
or hot-fix in filesystems has become useless, and consequently doesn't
exist in NSS, instead, NSS immediately deactivates a drive when an OS
noticeable write error occurs.

CU,
--
Massimo Rosen
Novell Product Support Forum Sysop
No emails please!
http://www.cfc-it.de
CU,
--
Massimo Rosen
Micro Focus Knowledge Partner
No emails please!
http://www.cfc-it.de
0 Likes
The opinions expressed above are the personal opinions of the authors, not of Micro Focus. By using this site, you accept the Terms of Use and Rules of Participation. Certain versions of content ("Material") accessible here may contain branding from Hewlett-Packard Company (now HP Inc.) and Hewlett Packard Enterprise Company. As of September 1, 2017, the Material is now offered by Micro Focus, a separately owned and operated company. Any reference to the HP and Hewlett Packard Enterprise/HPE marks is historical in nature, and the HP and Hewlett Packard Enterprise/HPE marks are the property of their respective owners.