OpenVMS utilities & system help

[WADE S]

I'm using OpenVMS 8.3-1H1 on a BL860c.
I'm not an OpenVMS user, and I've read several of the documents. I'm wondering if there are any easy utilities that will help me discover FC disk devices in a SAN, and generate I/O to them.
I also need to show my FC Mezz Card info, driver info, etc..
Can anyone help me out please?

Wade

[Robert Gezelter]

Wade,

I do not have a BL860 manual handy at this instant, but there should be instructions in that manual for associating WWIDs with fibre channel devices.

Talking this through using ITRC can get frustrating. I am not in my office at the moment, but I can be reached via email (address on my www site).

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

[WADE S]

Thank you for the help.

I'm currently reading about the VMS_BCFG utility.
Can you tell me if I have to go into the EFI utility, and configure the devices. I don't want to use them as boot devices, but I need to use them for data storage. Here's the procedure I found.

shell> cd fs0:\efi\vms\
fs0:\efi\vms\> vms_bcfg boot fibre 1
fs0:\efi\vms\> reconnect -r
â ¦
fs0:\efi\vms\> map -r -Fs

[Robert Gezelter · Accepted solution]

Wade,

No, VMS_BCFG (as documented in the manual, which is available online for other readers of this thread), is for setting up Bootable devices.

You need to configure the SAN switch to present the devices to the OpenVMS system. Then use SYSMAN IO AUTOCONFIGURE to bring them online to the OpenVMS system.

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

[WADE S]

I'm not sure which devices they are. All of the SAN config is done. Here's the display from the OVMS system.

Device Device Error
Name Status Count
HP$DKA100: Mounted 0 WA
HP$DNA0: Online wrtlck 0

Device Device Error
Name Status Count
OPA0: Online 0
FTA0: Offline 0
TTA0: Online 0
ASN0: Online 0
TNA0: Online 0
TNA2: Online 0

Device Device Error
Name Status Count
PKA0: Online 0
SRA0: Online 0
MPA0: Online 0
CMA0: Online 0
SEA0: Online 0
GHA0: Online 0
OHA0: Online 0
OHB0: Online 0
EHA0: Online 0
EWA0: Online 0
EWA2: Online 0
EWA3: Online 0
EWA4: Online 0
EWB0: Online 0
EWC0: Online 0
EWD0: Online 0
FGA0: Online 0
PGA0: Online 0
FGB0: Online 0
PGB0: Online 0
FGC0: Online 3
PGC0: Online 0
FGD0: Online 3
PGD0: Online 0
UCM0: Online 0
USB0: Online 0
HID0: Online 0
KBD0: Online 0
MOU0: Online 0
SMA0: Online 0
INA0: Offline 0
IMU0: Offline 0
IKU0: Offline 0
WSA0: Offline 0
WSA1: Online 0
PPP0: Online 0
PWIP0: Online 0
$

[Robert Gezelter]

Wade,

As a guess, check your SAN configuration as to which devices are presented to the OpenVMS system.

If your devices are configured, but they are not configured to be visible to the OpenVMS system, then the automatic IO AUTOCONFIGURE that is part of the startup will not find the devices.

- Bob Gezelter, http://www.rlgsc.com
P.S. I will be leaving my www browser shortly, continuing this is easier if we establish voice contact via email.

[WADE S]

I'll also be getting off of here until later this evening.
Thank you for the help you've provided so far.
The OpenVMS system should be discovering EVA storage. I've never even seen an OpenVMS system until last week, and using it is totally foreign to me.
I'll mess around with the autoconfigure later this evening.

Thanks, the help is appreciated.
Wade

[Robert Gezelter]

Wade,

You are welcome.

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

[Robert Gezelter]

Wade,

From your posted device display, my "guess" would be that the autoconfigure is working. However, I would suspect that the SAN is not configured to present the disk volumes to the OpenVMS system.

Please post the configuration information from the storage controller.

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

[WADE S]

Robert,

I have one LUN on an EVA3000 & one on an EVA8000 presented to the OVMS system. The fabric switch has 1 zone with both FC ports & 2 ports from the 3000, and 4 from the 8000. From doing a little searching, a SHOW DEVICE/FULL shows me that my FC ports are configured. I just have to figure out why the disks aren't showing up.

Device FGA0:, device type KGPSA Fibre Channel, is online, shareable, error
logging is enabled.

Error count 0 Operations completed 0
Owner process "" Owner UIC [SYSTEM]
Owner process ID 00000000 Dev Prot S:RWPL,O:RWPL,G,W
Reference count 2 Default buffer size 0
Current preferred CPU Id 0 Fastpath 1
Current Interrupt CPU Id 0
FC Port Name 1000-0000-C956-3E22 FC Node Name 2000-0000-C956-3E22

Device PGA0:, device type SCSI FCP, is online, error logging is enabled.

Error count 0 Operations completed 97
Owner process "" Owner UIC [SYSTEM]
Owner process ID 00000000 Dev Prot S:RWPL,O:RWPL,G,W
Reference count 0 Default buffer size 65535
Current preferred CPU Id 0 Fastpath 1

Device FGB0:, device type KGPSA Fibre Channel, is online, shareable, error
logging is enabled.

Error count 0 Operations completed 0
Owner process "" Owner UIC [SYSTEM]
Owner process ID 00000000 Dev Prot S:RWPL,O:RWPL,G,W
Reference count 2 Default buffer size 0
Current preferred CPU Id 1 Fastpath 1
Current Interrupt CPU Id 1
FC Port Name 1000-0000-C956-3E23 FC Node Name 2000-0000-C956-3E23

[Martin Vorlaender]

>>>
I'm not sure which devices they are. All of the SAN config is done. Here's the display from the OVMS system.

Device Device Error
Name Status Count
HP$DKA100: Mounted 0 WA
HP$DNA0: Online wrtlck 0
...
<<<

The FC disks should show up like $1$DGAn

If you don't see them, and if SYSMAN IO AUTOCONFIGURE doesn't discover them, I'd guess there's a problem in the presentation (i.e. OS Unit ID set to 0 on the EVA or VMS system HBA not in the same zone as the EVA on the FC switch).

Note that the EVA 3000 doen't present any vdisks when the cache batteries are not in a working order.

cu,
Martin

[Martin Vorlaender]

>>>
I'm not sure which devices they are. All of the SAN config is done. Here's the display from the OVMS system.

Device Device Error
Name Status Count
HP$DKA100: Mounted 0 WA
HP$DNA0: Online wrtlck 0
...
<<<

The FC disks should show up like $1$DGAn

If you don't see them, and if SYSMAN IO AUTOCONFIGURE doesn't discover them, I'd guess there's a problem in the presentation (e.g. OS Unit ID set to 0 on the EVA, or VMS system HBA not in the same zone as the EVA on the FC switch).

Note that the EVA 3000 doen't present any vdisks when the cache batteries are not in a working order.

cu,
Martin

[Robert Gezelter]

Wade,

Please post the information about the "disks" that the EVAs are presenting (from the EVA side), it was fairly clear that the OpenVMS ports were being found from the presence of the port controllers in the SHOW DEVICE output (as Martin has observed).

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

[Jan van den Ende]

Wade,

If the SAN config was done AFTER the VMS boot, then the SAN devices still have to be "discovered".

From a suitably priv'd account (CMKRNL,DIAG,LOG_IO,SYSPRV minimal) issue

$ MC SYSMAN IO FIND_WWID.

Now you SHOULD see (a) GG device(s).

If not, then the SAN config and/or connections are NOT Ok.
In that case (have SAN Management) make sure of LUN 0 (zero) NOT being in use, and zoning being set up correctly.

If you DO see GG device(s) an
$ MC SYSMAN IO AUTO should now find all SAN disks (and tapes) as DG (and MG) devices.

ps. for SAN -- boot -- device(s) only an extra WWIDMGR action at console mode is required.

hth.

Proost.

Have one on me.

jpe

[WADE S]

Alright, that may be the issue. The OS unit ID on both disks is set to 0. How do i know what it needs to be set to, or does anything other than 0 work?

[WADE S]

Alright, that may be the issue. The OS unit ID on both disks is set to 0. How do I know what it needs to be set to, or does anything other than 0 work?

[Art Wiens]

You can make the unit id's "anything" you want, but they must be unique across the fabric. Something reasonable like 100 and 101 will net you devices on the VMS system named $1$DGA100: and $1$DGA101: .

Cheers,
Art

[Jan van den Ende]

Wade,

>>>
The OS unit ID on both disks is set to 0.How do I know what it needs to be set to, or does anything other than 0 work?
<<<

1. NONE must be set to 0 (zero)
2. EACH must be set unique.
3. Depending on DEVICE_NAMING, they can be freely set from 1 tru 9999 or from 1 tru 32767.
Either setting, if you take your pick below 9999 you are GO.

hth

Proost.

Have one on me.

jpe

[Jan van den Ende]

Wade.

Upon rereading, let me get more precise.
It is the VMS system, which will assume the LUN 0 (zero), as seen by SAN.
Seen from VMS, that will be GGA#, with # going up from 1 for each additional SAN.
It will be the SAN LUNs, that will be used as the DGA#### device numbers as seen from VMS.

Proost.

Have one on me.

jpe

[WADE S]

OK, that did the trick.
Device Device Error Volume Free Trans Mnt
Name Status Count Label Blocks Count Cnt
HP$DKA100: Mounted 0 WADEOVMS 62132512 324 1
HP$DNA0: Online wrtlck 0
$1$DGA3001: (HP) Online 0
$1$DGA8001: (HP) Online 0

Now that I'm discovering the devices, what steps do I take to pass data to them?

Are there any tools available for OpenVMS, that allow me to generate lots of IO to the disks to test performance?

[Robert Gezelter]

Wade,

Probably one of the best abusers of disk speed is BACKUP.

I recommend that you INITIALIZE the DGA devices appropriately (see the HELP text), and then start with an BACKUP/IMAGE of your system device.

The System Manager's Manual will give useful introductory information for that and other tasks. It is available electronically on the OpenVMS www site at http://www.hp.com/go/OpenVMS

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

[Jan van den Ende]

Wade.

Bob wrote:
>>>
INITIALIZE the DGA devices appropriately
<<<

"appropriately" for SAN devices will at least be with
/CLUSTER=<16 or multiple thereoff)
/LIMIT=<1Tb>, ie, 2G blocks (to eanble later on-line growth)
/NOHIGHWATER (if that level of security were required, they should allow only VERY experienced Ssytem Managers; swithing it off is VERY favorable for performance)

After that, if you use BACKUP to load data, be sure to incluse /NOINIT in the BACKUP command.

hth

Proost.

Have one on me.

jpe

[Art Wiens]

There is a standard test package for VMS called UETP - User Environment Test Package:

http://h71000.www7.hp.com/doc/82FINAL/aa-pv5nj-tk/aa-pv5nj-tk.HTMl

One of the subsets of tests are "device" which will simulate user loads on devices it finds which have been prepared for testing. Have a read on how to set it up.

Art

[WADE S]

If I wanted to initialize $1$DGA3001, and then backup all system files to that disk,, what would the command syntax be?

& Is there a way to do it repeatedly?

[Jan van den Ende]

Wade,

>>>
If I wanted to initialize $1$DGA3001
<<<

Do you intebd to use shadowing?

For now, I will assume NO.

$ INIT $1$DGA3001 /NOHIGH/OWN=.../cluster=16/Share/limit=<2**31>
.. if you want to use lots of small files, add /DIRECTORIES and /HEADER, values larger than default
For /OWNER either SYSTEM or [1,1] are usual.

To roll a backup to this disk:

$ MOUNT /FOREIGN $1$DGA3001:
$ BACHUP/NOINIT/IMAGE $1$GDA3001:

$ Dismou $1$dga3001
$ Moun/system $1$dga3001

Success!

Proost.

Have one on me.

jpe