FYI: Odd PowerMac 2-socket/2-cores-each (16 GiByte) crashes: looks like over-temperature (a "quad" bites the dust?)

Mark Millard marklmi at yahoo.com
Mon Jul 20 04:28:05 UTC 2020


During my attempt to upgrade from head -r360311 to
-r363123 the old PowerMac has quit responding, without
notification of why. The fans were going (expected
even during normal operation for what I was doing)
but the fan(s) did not wind up to full-noise/speed.

Well, the machine finally did an automatic shutdown
successfully, so the logs files were updated. What I
found was:

Jul 19 19:03:26 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.5 C) exceeds critical temperature (90.0 C); count=1
Jul 19 19:03:28 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.2 C) exceeds critical temperature (90.0 C); count=2
Jul 19 19:08:41 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.3 C) exceeds critical temperature (90.0 C); count=1
Jul 19 19:08:43 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.3 C) exceeds critical temperature (90.0 C); count=2
Jul 19 19:08:45 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.3 C) exceeds critical temperature (90.0 C); count=3
Jul 19 19:08:49 FBSDG5L2 syslogd: last message repeated 1 times
Jul 19 19:08:51 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.2 C) exceeds critical temperature (90.0 C); count=4
Jul 19 19:12:11 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.2 C) exceeds critical temperature (90.0 C); count=1
Jul 19 19:14:06 FBSDG5L2 syslogd: last message repeated 1 times
Jul 19 19:14:12 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.7 C) exceeds critical temperature (90.0 C); count=1
Jul 19 19:14:14 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.7 C) exceeds critical temperature (90.0 C); count=2
Jul 19 19:14:16 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 91.0 C) exceeds critical temperature (90.0 C); count=3
Jul 19 19:14:18 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.7 C) exceeds critical temperature (90.0 C); count=4
Jul 19 19:14:28 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.5 C) exceeds critical temperature (90.0 C); count=2
Jul 19 19:14:30 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.8 C) exceeds critical temperature (90.0 C); count=3
Jul 19 19:14:32 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.8 C) exceeds critical temperature (90.0 C); count=4
Jul 19 19:14:34 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.7 C) exceeds critical temperature (90.0 C); count=5
Jul 19 19:14:36 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.3 C) exceeds critical temperature (90.0 C); count=6
Jul 19 19:14:36 FBSDG5L2 kernel: WARNING: CPU B1 DIODE TEMP temperature exceeded critical temperature 6 times in a row; shutting down!
Jul 19 19:14:38 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.2 C) exceeds critical temperature (90.0 C); count=7
Jul 19 19:14:38 FBSDG5L2 kernel: WARNING: CPU B1 DIODE TEMP temperature exceeded critical temperature 7 times in a row; shutting down!
Jul 19 19:14:39 FBSDG5L2 kernel: .
Jul 19 19:14:39 FBSDG5L2 ntpd[1153]: ntpd exiting on signal 15 (Terminated)
Jul 19 19:14:40 FBSDG5L2 kernel: .
Jul 19 19:14:42 FBSDG5L2 kernel: WARNING: Current temperature (CPU B1 DIODE TEMP: 90.5 C) exceeds critical temperature (90.0 C); count=7
Jul 19 19:14:42 FBSDG5L2 kernel: WARNING: CPU B1 DIODE TEMP temperature exceeded critical temperature 7 times in a row; shutting down!
Jul 19 19:14:43 FBSDG5L2 kernel: .
Jul 19 19:14:44 FBSDG5L2 syslogd: exiting on signal 15

It has a single-pump liquid cooling system.

I'm now testing the media in the 12 GiByte 2-socket/2-cores-each
(2 pumps) as a cross check on HW vs. SW. (Likely HW.) I've
started up poudriere again. The 16 GiByte PowerMac failed in
somewhat under 30 min of poudriere bulk activity (with a load
average of more than 4 much of the time). . . . On the 12 GiByte,
it has been 105 min+ with 0 temperature notices.

So another old single-pump "Quad" may have bit the dust but the
dual-pump one is still going.

The 12 GiByte PowerMac has bad memory slot(s) and so cannot have
16 GiBytes of usable RAM. (I no longer remember the details.)

===
Mark Millard
marklmi at yahoo.com
( dsl-only.net went
away in early 2018-Mar)



More information about the freebsd-ppc mailing list