StorageTek Virtual Storage Manager System 6 (VSM6)
 

PLA-Support>Sun Systems>TAPE>Virtual Tape>SN-TP: VSM6
 

The customer sees messages like the below on their MVS/MainFrame/Host console
N 4040000 SB90     2014118 16:26:26.17 JOB97304 00000010  IOS051I INTERFACE
TIMEOUT DETECTED ON 5648,BF,**,**02,PCHID=04A2
With in the VSM6 product we need to be able to map the above message to a FICON port to isolate the cause of the messages.

In this Document

Goal

Solution

References

Created from <SR 3-8965923147>

Applies to:

Goal

 Map the host message to a FICON port on the VSM6 that should lead to a quicker resolution if you need to engage engineering.

N 4040000 SB90     2014118 16:26:26.17 JOB97304 00000010  IOS051I INTERFACE
TIMEOUT DETECTED ON 5648,BF,**,**02,PCHID=04A2

Solution

 Below are the messages that the customer will see on the console and in syslog on the host.

The above host message IOS051I is broken down as follows:
IOS051I  - INTERFACE TIMEOUT DETECTED ON
dev,chp,cmd,stat,PCHID=pppp,EXTSS=ss

Explanation: The channel subsystem detected a timeout condition during the operation of device dev.
In the message text:
dev - The device number
chp - The channel path identifier (CHPID) if known; otherwise, this field is set to asterisks.
cmd - The failing command code, if known; otherwise, this field is set to asterisks.
stat - The device and subchannel status, if known; otherwise, this field is set to asterisks.
PCHID=pppp The physical channel id, if known.
EXTSS=ss The subchannel extended status, if known.

To decode the above MVS/HOST message follow below steps.
The device number example below, came from the first IOS message above:
IOS051I INTERFACE TIMEOUT DETECTED ON 5648,BF,**,**02,PCHID=04A2
Using the above decoder ring, 5648 is the customer device number, and BF is the chpid number.  Assuming his device number range for this VTSS is 5600-56FF, the last 2 digits of his device number are "control unit" and "device".  In VSM, we code these as Vc0d, where c=control unit and d=device, hence you get V408.

Now looking at a current state save from the above system we can map that as follows:
1. Identify the chpid in question decoded from above, we need to know which fd_user.####.log file the chpid's are in.
    We use the below grep command to find the correct log file and process ID.

2. We use the below command to find the FICON port in question, mapping the Process_ID from above to the correct port number.

3. Then if you look for abort log messages in the tikka.out.p# port logs, and match up the time and device numbers, then you can be certain.

 From the above we can see we need to focus on p1 in tikka.out.p1.12 file and the port in question is the bottom port on the left hand FICON card looking at the back of the system.

References

This solution has no attachment