Hi Rouven,
On Wed, 02 May 2007 23:33:58 +0200, Rouven Sacha wrote:
> I'm new to the list, this is my first post, so please tell me if there's
> any information missing.
>
> After a reaching 75?C high, our hosted server got a forced shutdown due
> to an acpi event today.
>
> May 2 18:41:37 blinkenmail kernel: ACPI: Critical trip point
> May 2 18:41:37 blinkenmail kernel: Critical temperature reached (76 C),
> shutting down.
> May 2 18:41:37 blinkenmail shutdown[2522]: shutting down for system
> halt
> May 2 18:41:37 blinkenmail kernel: ACPI: Unable to turn cooling device
> [c1b17dec] 'on'
> May 2 18:41:38 blinkenmail init: Switching to runlevel: 0
> May 2 18:41:43 blinkenmail kernel: ACPI: Critical trip point
> May 2 18:41:43 blinkenmail kernel: Critical temperature reached (75 C),
> shutting down.
> May 2 18:41:43 blinkenmail kernel: ACPI: Unable to turn cooling device
> [c1b17dec] 'on'
You may want to report this to the ACPI folks and/or your hardware
vendor, as this looks like a bug either in the Linux acpi code, or in
your system's BIOS.
> It's a debian etch xen-dom0 running linux 2.6.18-4. I tried to
> investigate, so I installed lm-sensors ( 2.10.1-3).
>
> sensors-detect:
>
> Driver `eeprom' (should be inserted):
> Detects correctly:
> * Bus `SMBus Via Pro adapter at 5000'
> Busdriver `i2c-viapro', I2C address 0x50
> Chip `eeprom' (confidence: 6)
> * Bus `SMBus Via Pro adapter at 5000'
> Busdriver `i2c-viapro', I2C address 0x51
> Chip `eeprom' (confidence: 6)
>
> Driver `w83627hf' (should be inserted):
> Detects correctly:
> * ISA bus address 0x0290 (Busdriver `i2c-isa')
> Chip `Winbond W83697HF Super IO Sensors' (confidence: 9)
>
> modprobed: i2c-viapro eeprom w83627hf
>
> 23:29:57 blinkenmail:~ # lsmod
> Module Size Used by
> i2c_dev 9316 0
> tun 11104 1
> ipv6 229088 34
> dm_snapshot 16320 0
> dm_mirror 20048 0
> dm_mod 51000 2 dm_snapshot,dm_mirror
> w83627hf 23344 0
> hwmon_vid 3552 1 w83627hf
> eeprom 7792 0
> i2c_isa 5920 1 w83627hf
> i2c_viapro 9012 0
> i2c_core 20480 5 i2c_dev,w83627hf,eeprom,i2c_isa,i2c_viapro
> tulip 48768 0
> shpchp 33632 0
> pci_hotplug 29472 1 shpchp
> psmouse 35880 0
> pcspkr 3840 0
> serio_raw 7428 0
> via_agp 10432 1
> agpgart 32264 1 via_agp
> evdev 9856 0
> rtc 13300 0
> ext3 120072 1
> jbd 53224 1 ext3
> mbcache 9124 1 ext3
> ide_disk 15712 3
> via82cxxx 9156 0 [permanent]
> generic 6244 0 [permanent]
> ide_core 112392 3 ide_disk,via82cxxx,generic
> via_rhine 24456 0
> mii 6112 1 via_rhine
> ehci_hcd 29288 0
> uhci_hcd 22188 0
> sata_via 10980 0
> usbcore 114372 3 ehci_hcd,uhci_hcd
> libata 90868 1 sata_via
> scsi_mod 125160 1 libata
> thermal 14376 0
> processor 29608 1 thermal
> fan 5572 0
>
>
> and sensors gave me:
>
> 21:01:27 blinkenmail:plugins # sensors
> w83697hf-isa-0290
> Adapter: ISA adapter
> VCore: +1.68 V (min = +0.26 V, max = +0.54 V) ALARM
> +3.3V: +3.26 V (min = +0.66 V, max = +3.87 V)
> +5V: +5.03 V (min = +0.03 V, max = +0.86 V) ALARM
> +12V: +11.49 V (min = +5.72 V, max = +4.99 V) ALARM
> -12V: -11.62 V (min = -13.59 V, max = -3.07 V)
> -5V: -4.90 V (min = +0.53 V, max = -2.89 V) ALARM
> V5SB: +5.43 V (min = +5.46 V, max = +1.29 V) ALARM
> VBat: +3.28 V (min = +0.14 V, max = +1.04 V) ALARM
> fan1: 2220 RPM (min = 4963 RPM, div = 8) ALARM
> fan2: 4017 RPM (min = 2481 RPM, div = 8)
> temp1: +28?C (high = +64?C, hyst = -102?C) sensor = thermistor
> temp2: +68.5?C (high = +75?C, hyst = +70?C) sensor = diode
> alarms: Chassis intrusion detection ALARM
> beep_enable:
> Sound alarm enabled
>
> The system had a load of under 0.10 and the values for temp1 changed
> between 65 and 69?C. I told the support guy of our hoster that there
> might be something wrong with the cpu fan and he replaced it with a new
> one, he said.
It seems he really did, as the fan1 reading is higher now (2556 RPM
instead of 2220 RPM).
> Now sensors gives me a constant value of -93,5?C for temp2:
>
> 23:24:13 blinkenmail:~ # sensors
> w83697hf-isa-0290
> Adapter: ISA adapter
> VCore: +1.68 V (min = +0.26 V, max = +0.54 V) ALARM
> +3.3V: +3.25 V (min = +0.66 V, max = +3.89 V)
> +5V: +5.03 V (min = +0.03 V, max = +0.86 V) ALARM
> +12V: +11.55 V (min = +5.59 V, max = +4.99 V) ALARM
> -12V: -11.54 V (min = -14.91 V, max = -3.07 V)
> -5V: -4.90 V (min = +0.53 V, max = -6.10 V) ALARM
> V5SB: +5.46 V (min = +5.46 V, max = +1.29 V) ALARM
> VBat: +3.28 V (min = +0.14 V, max = +1.04 V) ALARM
> fan1: 2556 RPM (min = 4963 RPM, div = 8) ALARM
> fan2: 4017 RPM (min = 2481 RPM, div = 8)
> temp1: +33?C (high = +64?C, hyst = -102?C) sensor = thermistor
> temp2: -93.5?C (high = +75?C, hyst = +70?C) sensor = diode
> alarms: Chassis intrusion detection ALARM
> beep_enable:
> Sound alarm enabled
>
>
> What does that mean?
>
> Any hints are appreciated,
I don't know. Did the temp2 value change at all after that? If this is
the first time you installed lm_sensors on this machine, we can't
exclude that the fan change and the temperature value breakage are
uncorrelated. Maybe it would have happened anyway. Although I admit
this is a suspicious coincidence.
I guess that you didn't power-off the system, and that it isn't really
an option?
Does the ACPI temperature look reasonable ("acpi -t" or look directly
in /proc/acpi/thermal_zone/*/temperature)? It could be a conflict
between ACPI and the w83627hf driver. Can you please send me (in
private) a copy of /proc/acpi/dsdt?
--
Jean Delvare
