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/
6MDC1141_CONSOLES.txt
Description: Binary data
MQ-Guest-IP-Config.txt
Description: Binary data
Kernel inof.txt
Description: Binary data
