From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ben Kamen Date: Sun, 13 May 2012 23:57:36 +0000 Subject: Re: [lm-sensors] Supermicro X7SPA-HF: in4 ALARM Message-Id: <4FB04A70.2080303@benjammin.net> List-Id: References: <01fa01cd1430$fed0b990$fc722cb0$@lucidpixels.com> In-Reply-To: <01fa01cd1430$fed0b990$fc722cb0$@lucidpixels.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable To: lm-sensors@vger.kernel.org Ok, after looking around some more -- I should be talking though ipmisensor= s and not directly communicating with the w83627ehf directly. it finds the ipmi interface: > Found `IPMI BMC KCS' at 0xca2... Success! > (confidence 8, driver `ipmisensors') > but then tells me: > Warning: the required module ipmisensors is not currently installed > on your system. If it is built into the kernel then it's OK. > Otherwise, check http://www.lm-sensors.org/wiki/Devices for > driver availability. and I'm using the RHEL 6 released version of lm-sensors: lm_sensors.x86_64 3.1.1-10.el6 @base So where do I get ipmisensors? The wiki says I should use ipmitool -- but t= hen how does that integrate? Thanks, -Ben On 05/13/12 5:30 PM, Ben Kamen wrote: > Hey guys, in response to this email below from Jean, > > Can we still use lm_sensors on the winbond IC on the board or should lm_s= ensors be disabled entirely? > > I'm also using an X7SPA-HF and the thermal management goes wonky after a = few hours of normal operation. > > (temp alarm, fans report erroneous speeds and so forth) > > When this started happening, I began to wonder about contention between t= he IPMI controller onboard and what linux might be trying to do at the same= time. > > Thanks, > > -Ben > >> On Fri, 6 Apr 2012 16:08:12 -0400, Justin Piszcz wrote: >>> Hello, >>> >>> Was curious why the max voltage does not show up properly for in4 (VDIM= M) >>> for this X7SPA-HF using lm sensors 3.3.1 or 3.3.2? >>> >>> On the sensor reading page for IPMI: >>> VDIMM Normal 1.84 Volts >> >> You should not mix IPMI with native Linux hardware monitoring driver. >> There is no mutual exclusion on device access, and havoc can (and will) >> happen. >> >>> This is with lm_sensors: >>> >>> -- >>> >>> $ sensors >>> w83627dhg-isa-0ca0 >>> Adapter: ISA adapter >>> Vcore: +1.16 V (min =3D +0.72 V, max =3D +1.39 V) >>> Vnbcore: +1.04 V (min =3D +0.94 V, max =3D +1.16 V) >>> AVCC: +3.34 V (min =3D +2.98 V, max =3D +3.63 V) >>> +3.3V: +3.34 V (min =3D +2.98 V, max =3D +3.63 V) >>> VDIMM: +1.84 V (min =3D +1.62 V, max =3D +0.00 V) ALARM >>> +5V: +1.25 V (min =3D +1.13 V, max =3D +1.38 V) >>> +12V: +0.75 V (min =3D +0.67 V, max =3D +0.83 V) >>> 3VSB: +3.30 V (min =3D +2.98 V, max =3D +3.63 V) >>> Vbat: +3.06 V (min =3D +2.70 V, max =3D +3.30 V) >>> SYS Temp: +42.0=B0C (high =3D +75.0=B0C, hyst =3D +70.0=B0C) sensor =3D= thermistor >>> CPU Temp: +40.5=B0C (high =3D +90.0=B0C, hyst =3D +87.0=B0C) sensor =3D= diode >>> Case Temp: +24.0=B0C (high =3D +80.0=B0C, hyst =3D +75.0=B0C) sensor = =3D diode >>> $ sensors --version >>> sensors version 3.3.1 with libsensors version 3.3.1 >>> >>> lm_sensors-3.3.2$ prog/sensors/sensors --version >>> sensors version 3.3.2 with libsensors version 3.3.1 >>> >>> w83627dhg-isa-0ca0 >>> Adapter: ISA adapter >>> Vcore: +1.16 V (min =3D +0.72 V, max =3D +1.39 V) >>> Vnbcore: +1.04 V (min =3D +0.94 V, max =3D +1.16 V) >>> AVCC: +3.34 V (min =3D +2.98 V, max =3D +3.63 V) >>> +3.3V: +3.34 V (min =3D +2.98 V, max =3D +3.63 V) >>> VDIMM: +1.84 V (min =3D +1.62 V, max =3D +0.00 V) ALARM >>> +5V: +1.25 V (min =3D +1.13 V, max =3D +1.38 V) >>> +12V: +0.75 V (min =3D +0.67 V, max =3D +0.83 V) >>> 3VSB: +3.30 V (min =3D +2.98 V, max =3D +3.63 V) >>> Vbat: +3.06 V (min =3D +2.70 V, max =3D +3.30 V) >>> SYS Temp: +42.0=B0C (high =3D +75.0=B0C, hyst =3D +70.0=B0C) sensor =3D= thermistor >>> CPU Temp: +40.5=B0C (high =3D +90.0=B0C, hyst =3D +87.0=B0C) sensor =3D= CPU diode >>> Case Temp: +24.5=B0C (high =3D +80.0=B0C, hyst =3D +75.0=B0C) sensor = =3D CPU diode >>> >>> -- >>> >>> I also tried hardcoding in4_{min,max}: >>> >>> $ cat /etc/sensors3.conf >>> >>> chip "w83627ehf-*" "w83627dhg-*" "w83667hg-*" "nct6775-*" "nct6776-*" >>> >>> label in0 "Vcore" >>> label in1 "Vnbcore" >>> label in2 "AVCC" >>> label in3 "+3.3V" >>> label in4 "VDIMM" >>> label in5 "+5V" >>> label in6 "+12V" >>> label in7 "3VSB" >>> label in8 "Vbat" >>> label temp1 "SYS Temp" >>> label temp2 "CPU Temp" >>> label temp3 "Case Temp" >>> >>> set in2_min 3.3 * 0.90 >>> set in2_max 3.3 * 1.10 >>> set in3_min 3.3 * 0.90 >>> set in3_max 3.3 * 1.10 >>> # http://lists.lm-sensors.org/pipermail/lm-sensors/2010-November/030368= .html >>> set in4_min 1.62 >>> set in4_max 1.98 >>> set in7_min 3.3 * 0.90 >>> set in7_max 3.3 * 1.10 >>> set in8_min 3.0 * 0.90 >>> set in8_max 3.0 * 1.10 >>> >>> ignore fan1 >>> ignore fan2 >>> ignore fan3 >>> ignore fan4 >>> ignore fan5 >>> ignore cpu0_vid >>> ignore intrusion0 >> >> Did you run "sensors -s" after changing the configuration file? >> >> -- >> Jean Delvare > > _______________________________________________ > lm-sensors mailing list > lm-sensors@lm-sensors.org > http://lists.lm-sensors.org/mailman/listinfo/lm-sensors --=20 Ben Kamen - O.D.T., S.P. ---------------------------------------------------------------------- eMail: ben@benjammin.net http://www.benjammin.net http://www.linkedin.com/in/benkamen Fortune says: Only a fool fights in a burning house. -- Kank the Klingon, "Day of the Dove", stardate unknown - - NOTICE: All legal disclaimers sent to benjammin.net/benkamen.net or any of it's affiliated domains are rendered null and void on receipt of communications will be handled/considered as such. _______________________________________________ lm-sensors mailing list lm-sensors@lm-sensors.org http://lists.lm-sensors.org/mailman/listinfo/lm-sensors