Below you will find a description, along with attachments, of a perceived
HiperSocket issue we had with one of our agencies on Sunday, 8-7-11. This
is for the most part the same text I submitted to IBM support for potential
problem identification (PMR - 00129,MME,000 - IBM Confidential). Support
was not able to offer additional information or resolution given the
limited amount of diagnostic information available and that we were able to
provide.
 I am calling on your collective expertise, experiences, and history trying
to find any known issues, known causes, known fixes, circumventions etc
etc.
We need to, as best we can, identify what could create this scenario and
how can we avoid it in the furture. The account did incur an SLA violation
as a result  of the incident which makes this very visible from our
perspective.

Problem Description:

 Several weeks ago the COPA account IPL'ed one of our z/VM LPARs(z/VM 5.4) 
which hosts
 many
 z/Linux guests. One of the z/Linux guests(SLES 10 SP3), an MQ Server,
 uses a Hipersocket interface to communicate with a z/OS LPAR on the same CPU. 
At the
 z/Linux boot
 up, we had a failure where the virtual address DEDICATEd in the VM directory 
to the
 physical addresses associated with the HiperSocket IQD channel was not 
available, or not
 detected, to the z/Linux Network Configuration BOOT sequence. The HSI0 
interface did not
 start and returned the following BOOT message:

 Setting up network interfaces:
     lo
     lo        IP address: 127.0.0.1/8
               IP address: 127.0.0.2/8
 Checking for network time protocol daemon (NTPD): ..unused
     lo
 [1A..done
     eth0
     eth0      configuration: qeth-bus-ccw-0.0.0900
     eth0      IP address: 10.26.6.141/25
 Checking for network time protocol daemon (NTPD): ..unused
     eth0
 [1A..done
 Waiting for mandatory devices:  hsi-bus-ccw-0.0.7000 __NSC__
 18
 17
 16
 15
 14
 13
 12
 11
 10
 9
 8
 7
 6
 5
 4
 3
 2
 1
 0
     hsi-bus-ccw-0.0.7000                No interface found
 [1A..failedSetting up service
 network  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .  ...failed
 Starting syslog services
 ..done

 This configuration has never failed in two years with an IPL every Sunday.
 The problem has not occurred since
 We confirmed that the DEDICATE(virtual to real) on z/VM was active
 and available. When the z/Linux guest was configuring its network
 interfaces it did not recognize the virtual address. There were no
 errors when the guest signed on to z/VM. The shut down and boot log for
 the guest is attached along with the NW config file. I marked the
 start of the HSI0 start up failure with "****************** FAILED
 STARTUP START".
 Subsequent to the interface failure at boot time, a HW UP command
 was issued for the HSI0 interface and it worked fine with no other
 action taken.

 Below are the guest boot log and network config files.

 (See attached file: 6MDC1141_CONSOLES.txt)(See attached file: 
MQ-Guest-IP-Config.txt)
 (See attached file: Kernel inof.txt)




Regards.


John R. Stout
IT Specialist, Server, Storage, and Data Management
IBM Global Tech Services
717-526-1239 (T/L 239-1239)
pager: 1-800-759-8888 pin 1740690
[email protected]

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
----------------------------------------------------------------------
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Attachment: 6MDC1141_CONSOLES.txt
Description: Binary data

Attachment: MQ-Guest-IP-Config.txt
Description: Binary data

Attachment: Kernel inof.txt
Description: Binary data

Reply via email to