Mark,
Since I responded, I had him do a complete trace of the ipl until it
started looping. I have already reviewed the trace and found that not
long after IPL, there was an invalid instruction.
After asking more questions, I found out that he was doing something I
did not know about.
The restored system was being tested not on their current CPU, but the
one they have in their DR site. The invalid instruction is valid on
their production machine, but is invalid on their DR machine. Z114 vs Z10.
Sometimes I just want to reach though the phone lines and see if I can
relocate their head to a lower orifice.
Tony Thigpen
Mark Post wrote on 2/25/22 17:13:
On 2/25/22 16:40, Tony Thigpen wrote:
Mark,
I did not think of tracing. Yes, it is in a tight loop. Here is a bit of
the log:
-> 00000000000039D0 STMG EB0F02000024 >> 0000000000000200 CC 0
00000000000039D6 AGHI A7FBFF38 CC 1
00000000000039DA LARL C0E0FFFFF9FB CC 1
00000000000039E0 BASR 0DEE -> 0000000000002DD0 CC 1
-> 0000000000002DD0 STMG EBDFF0880024 >> FFFFFFFFFB274FB8 CC 1
** 0000000000002DD0 PROG 0005 -> 00000000000039D0
ADDRESSING
-> 00000000000039D0 STMG EB0F02000024 >> 0000000000000200 CC 0
00000000000039D6 AGHI A7FBFF38 CC 1
00000000000039DA LARL C0E0FFFFF9FB CC 1
00000000000039E0 BASR 0DEE -> 0000000000002DD0 CC 1
-> 0000000000002DD0 STMG EBDFF0880024 >> FFFFFFFFFB274EF0 CC 1
** 0000000000002DD0 PROG 0005 -> 00000000000039D0
ADDRESSING
-> 00000000000039D0 STMG EB0F02000024 >> 0000000000000200 CC 0
00000000000039D6 AGHI A7FBFF38 CC 1
00000000000039DA LARL C0E0FFFFF9FB CC 1
00000000000039E0 BASR 0DEE -> 0000000000002DD0 CC 1
-> 0000000000002DD0 STMG EBDFF0880024 >> FFFFFFFFFB274E28 CC 1
** 0000000000002DD0 PROG 0005 -> 00000000000039D0
ADDRESSING
Yeah, that's going to be a pain to debug. The System.map file in /boot
would indicate what routine that's in, but after that it's most likely
going to require a kernel expert to figure out. You can tell that the
STMG instruction is trying to address what appears to be a negative (or
extremely large) address and getting an addressing exception. But how it
got there is a much bigger problem to figure out.
What version of SLES is this? What model machine is this running on? I'm
assuming it's the same machine as before, and just the DASD was swapped
out, but you never know.
Mark Post
----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or
visit
http://www2.marist.edu/htbin/wlvindex?LINUX-390
----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www2.marist.edu/htbin/wlvindex?LINUX-390
