IPFire Sensors <-> Supermicro IPMI

Martin
Posts: 11
Joined: May 20th, 2010, 2:09 pm

IPFire Sensors <-> Supermicro IPMI

Post by Martin » December 17th, 2013, 3:33 pm

Hallo Zusammen,

ich hätte mal ein kleines Problem.
Wir nutzen als Hardware Supermicro mit IPMI.
Die Supermicro hat als Mainboard, das X7SPE-HF-D525.
Wir nutzen keinen aktiven Lüfter in dem System.

Nun ist es so das die Hardware alle 2 - 3 Tage das Piepen anfängt.
Ich konnte den Fehler nun darauf eingrenzen, dass sich die IPMI Sensoren mit dem lmsensors der IPFire beißt.

In diversen Foren konnte ich lesen, dass andere das Problem gelöst haben in dem Sie in der "/etc/collectd.conf" den Punkt "load Plugin sensors" auskommentiert haben.
Leider jedoch verschwindet die Auskommentierung nach jeden neustart.

Vorerst lasse ich die Firewall jeden Tag neustarten, dann tritt dieses Problem nicht auf.

Hat hier evtl. jemand eine Idee wie ich das lösen könnte?

Danke schon mal!

Gruß Martin

Martin
Posts: 11
Joined: May 20th, 2010, 2:09 pm

Re: IPFire Sensors <-> Supermicro IPMI

Post by Martin » December 23rd, 2013, 9:38 am

Hallo Miteinander,

ich konnte nun eine Lösung finden.

In der /etc/init.d/collectd folgende Zeile (Markierung) auskommentieren:

# Enable sensors plugin if sensors found
if [ "$( sensors 2>&1 | grep 'No sensors found!' | wc -l )" == "1" ];
then
#sed -i -e "s|^LoadPlugin sensors|#LoadPlugin sensors|g" /etc/collectd.conf
else
sed -i -e "s|^#LoadPlugin sensors|LoadPlugin sensors|g" /etc/collectd.conf
fi



Zusätzlich in der /etc/collectd.conf folgende Zeile auskommentieren:
load Plugin sensor


Nun tritt das Piepen nicht mehr auf.

Schöne Weihnachten und einen Guten Rutsch ins neue Jahr.

Gruß Martin

User avatar
tam
Posts: 42
Joined: December 6th, 2007, 8:10 am

Re: IPFire Sensors <-> Supermicro IPMI

Post by tam » August 17th, 2015, 1:24 pm

Hi, ich habe das gleiche Problem. Ist das nach wie vor die "offizielle" Lösung?

Mainboard: SuperMicro X8SIL-F
IPFire 2.17 (i586) - Core Update 92

morlix
Community Developer
Community Developer
Posts: 175
Joined: July 18th, 2011, 10:08 am
Location: Stuttgart

Re: IPFire Sensors <-> Supermicro IPMI

Post by morlix » August 18th, 2015, 7:51 am

Hier wäre es mal interessant den Output von

Code: Select all

ipmitool sensor


zu sehen , während und vor dem piepsen. Ich vermute hier eher fehlerhafte bzw. nicht passende thresholds.

ipmitool ist aktuell noch nicht in der Distribution, ich könnte aber ein Binary dafür zur Verfügung stellen, wenn jemand testen will.

Gruß morlix

User avatar
tam
Posts: 42
Joined: December 6th, 2007, 8:10 am

Re: IPFire Sensors <-> Supermicro IPMI

Post by tam » August 18th, 2015, 8:00 am

Würde es testen, wenn ich die binary habe. Habe jetzt aber erst mal über Nacht neu gestartet, das Piepsen hat hier alle ganz nervös gemacht :-)

morlix
Community Developer
Community Developer
Posts: 175
Joined: July 18th, 2011, 10:08 am
Location: Stuttgart

Re: IPFire Sensors <-> Supermicro IPMI

Post by morlix » August 20th, 2015, 7:43 am

Hier das ipmitool Binary.

Keine Ahnung ob das so auf deiner IPFire läuft oder noch einige andere Dateien benötigt werden, aber teste es einfach mal. (.txt entfernen :-) )

Gruß morlix
Attachments
ipmitool.txt
(775.2 KiB) Downloaded 493 times

User avatar
tam
Posts: 42
Joined: December 6th, 2007, 8:10 am

Re: IPFire Sensors <-> Supermicro IPMI

Post by tam » August 20th, 2015, 7:48 am

Gleich mal ausprobiert:

Code: Select all

[root@ipfire ~]# ./ipmitool sensor
Could not open device at /dev/ipmi0 or /dev/ipmi/0 or /dev/ipmidev/0: No such file or directory


Die Module laden:

Code: Select all

modprobe ipmi_devintf
modprobe ipmi_si


Code: Select all

[root@ipfire ~]# ./ipmitool sensor
System Temp      | 33.000     | degrees C  | ok    | -9.000    | -7.000    | -5.000    | 75.000    | 77.000    | 79.000
CPU Temp         | 0x1        | discrete   | 0x0000| na        | na        | na        | na        | na        | na
FAN 1            | 955.000    | RPM        | ok    | 400.000   | 585.000   | 770.000   | 29260.000 | 29815.000 | 30370.000
FAN 2            | -2560.000  | RPM        | nr    | 400.000   | 585.000   | 770.000   | 29260.000 | 29815.000 | 30370.000
FAN 3            | -2560.000  | RPM        | nr    | 400.000   | 585.000   | 770.000   | 29260.000 | 29815.000 | 30370.000
FAN 4            | -2560.000  | RPM        | nr    | 400.000   | 585.000   | 770.000   | 29260.000 | 29815.000 | 30370.000
FAN 5            | -2560.000  | RPM        | nr    | 400.000   | 585.000   | 770.000   | 29260.000 | 29815.000 | 30370.000
CPU Vcore        | 0.936      | Volts      | ok    | 0.640     | 0.664     | 0.688     | 1.344     | 1.408     | 1.472
+3.3VCC          | 3.264      | Volts      | ok    | 2.816     | 2.880     | 2.944     | 3.584     | 3.648     | 3.712
+12 V            | 12.243     | Volts      | ok    | 10.494    | 10.600    | 10.706    | 13.091    | 13.197    | 13.303
CPU DIMM         | 1.528      | Volts      | ok    | 1.152     | 1.216     | 1.280     | 1.760     | 1.776     | 1.792
+5 V             | 4.992      | Volts      | ok    | 4.096     | 4.320     | 4.576     | 5.344     | 5.600     | 5.632
-12 V            | -12.680    | Volts      | ok    | -13.844   | -13.650   | -13.456   | -11.710   | -11.516   | -11.322
VBAT             | 3.120      | Volts      | ok    | 2.816     | 2.880     | 2.944     | 3.584     | 3.648     | 3.712
+3.3VSB          | 3.248      | Volts      | ok    | 2.816     | 2.880     | 2.944     | 3.584     | 3.648     | 3.712
AVCC             | 3.264      | Volts      | ok    | 2.816     | 2.880     | 2.944     | 3.584     | 3.648     | 3.712
Chassis Intru    | 0x1        | discrete   | 0x0100| na        | na        | na        | na        | na        | na
PS Status        | 0x1        | discrete   | 0x01ff| na        | na        | na        | na        | na        | na

morlix
Community Developer
Community Developer
Posts: 175
Joined: July 18th, 2011, 10:08 am
Location: Stuttgart

Re: IPFire Sensors <-> Supermicro IPMI

Post by morlix » August 21st, 2015, 8:12 am

Naja mit diesen

FAN 2 | -2560.000 | RPM | nr | 400.000 | 585.000 | 770.000 | 29260.000 | 29815.000 | 30370.000
FAN 3 | -2560.000 | RPM | nr | 400.000 | 585.000 | 770.000 | 29260.000 | 29815.000 | 30370.000
FAN 4 | -2560.000 | RPM | nr | 400.000 | 585.000 | 770.000 | 29260.000 | 29815.000 | 30370.000
FAN 5 | -2560.000 | RPM | nr | 400.000 | 585.000 | 770.000 | 29260.000 | 29815.000 | 30370.000


NR (Non-Recoverable) entries wundert es mich nicht, das der BMC irgendwann einen nervigen Warnton ausgibt.
Stimmt es, dass FAN 2- 5 nicht angeschlossen sind?

Gruß morlix

User avatar
tam
Posts: 42
Joined: December 6th, 2007, 8:10 am

Re: IPFire Sensors <-> Supermicro IPMI

Post by tam » August 24th, 2015, 6:54 am

Stimmt, die sind nicht angeschlossen.

Habe jetzt noch mal ein ipmitool sensor während dem Piepen gemacht, aber außer der System Temp, die von 33 auf 34°C ist, hat sich nichts geändert.

morlix
Community Developer
Community Developer
Posts: 175
Joined: July 18th, 2011, 10:08 am
Location: Stuttgart

Re: IPFire Sensors <-> Supermicro IPMI

Post by morlix » August 24th, 2015, 4:07 pm

Dein BMC meckert einfach die NR (Non-Recoverable) Errors an. Sprich die Lüfter welche nicht angeschlossen sind.
Die BMC's sind in dieser Hinsicht oft nicht gerade flexibel und erwarten das alles passt.

Gruß morlix

User avatar
tam
Posts: 42
Joined: December 6th, 2007, 8:10 am

Re: IPFire Sensors <-> Supermicro IPMI

Post by tam » August 24th, 2015, 4:10 pm

Auf der Kiste war vorher ein Windows Server, da hat nie was gepiept, also muss es ja doch irgendwie eine Software Geschichte sein, meinst Du nicht auch?

morlix
Community Developer
Community Developer
Posts: 175
Joined: July 18th, 2011, 10:08 am
Location: Stuttgart

Re: IPFire Sensors <-> Supermicro IPMI

Post by morlix » August 26th, 2015, 6:03 am

Hmm, da muss ich dir Recht geben, das ist strange. Meiner Erfahrung nach war hier immer das BMC bzw. dessen flexibilität ein Problem.

Lass uns mal folgendes versuchen. Lösch den SEL (System Event Log) des BMC mit "ipmitool sel clear" und wenn dann das piepen wieder kommt schauen wir uns das SEL an.
Entweder mit "ipmitool sel list" oder die ausführliche Variante mit "ipmitool sel elist".

Vielleicht bekommen wir daraus ein paar nützliche Informationen.
Das Problem muss imho von der BMC kommen, da diese den System Beeper steuert. Sprich es gilt "nur" herauszufinden was den BMC nach X Stunden oder Events veranlasst jetzt loszulegen.

Gruß morlix

User avatar
tam
Posts: 42
Joined: December 6th, 2007, 8:10 am

Re: IPFire Sensors <-> Supermicro IPMI

Post by tam » August 26th, 2015, 7:04 am

Hi morlix, super Idee mit dem elist.

Hier mal die Liste der events im August:

Code: Select all

1a | 08/03/2015 | 18:38:50 | Fan FAN 2 | Lower Non-critical going low  | Asserted | Reading -2560 < Threshold 770 RPM
1b | 08/03/2015 | 18:38:50 | Fan FAN 2 | Lower Critical going low  | Asserted | Reading -2560 < Threshold 585 RPM
1c | 08/03/2015 | 18:38:50 | Fan FAN 2 | Lower Non-recoverable going low  | Asserted | Reading -2560 < Threshold 400 RPM
1d | 08/05/2015 | 06:00:31 | Fan FAN 5 | Lower Non-critical going low  | Asserted | Reading -2560 < Threshold 770 RPM
1e | 08/05/2015 | 06:00:31 | Fan FAN 5 | Lower Critical going low  | Asserted | Reading -2560 < Threshold 585 RPM
1f | 08/05/2015 | 06:00:31 | Fan FAN 5 | Lower Non-recoverable going low  | Asserted | Reading -2560 < Threshold 400 RPM
20 | 08/05/2015 | 06:16:04 | Fan FAN 3 | Lower Non-critical going low  | Asserted | Reading -2560 < Threshold 770 RPM
21 | 08/05/2015 | 06:16:04 | Fan FAN 3 | Lower Critical going low  | Asserted | Reading -2560 < Threshold 585 RPM
22 | 08/05/2015 | 06:16:04 | Fan FAN 3 | Lower Non-recoverable going low  | Asserted | Reading -2560 < Threshold 400 RPM
23 | 08/05/2015 | 12:59:43 | Fan FAN 4 | Lower Non-critical going low  | Asserted | Reading -2560 < Threshold 770 RPM
24 | 08/05/2015 | 12:59:43 | Fan FAN 4 | Lower Critical going low  | Asserted | Reading -2560 < Threshold 585 RPM
25 | 08/05/2015 | 12:59:43 | Fan FAN 4 | Lower Non-recoverable going low  | Asserted | Reading -2560 < Threshold 400 RPM
26 | 08/06/2015 | 21:35:47 | Fan FAN 1 | Lower Non-critical going low  | Asserted | Reading -2560 < Threshold 770 RPM
27 | 08/06/2015 | 21:35:47 | Fan FAN 1 | Lower Critical going low  | Asserted | Reading -2560 < Threshold 585 RPM
28 | 08/06/2015 | 21:35:47 | Fan FAN 1 | Lower Non-recoverable going low  | Asserted | Reading -2560 < Threshold 400 RPM
29 | 08/06/2015 | 21:36:46 | Fan FAN 1 | Lower Non-recoverable going low  | Deasserted | Reading 955 < Threshold 400 RPM
2a | 08/06/2015 | 21:36:46 | Fan FAN 1 | Lower Critical going low  | Deasserted | Reading 955 < Threshold 585 RPM
2b | 08/08/2015 | 06:59:07 | Physical Security Chassis Intru | General Chassis intrusion () | Asserted
2c | 08/08/2015 | 07:29:38 | Fan FAN 2 | Lower Non-critical going low  | Asserted | Reading -2560 < Threshold 770 RPM
2d | 08/08/2015 | 07:29:38 | Fan FAN 2 | Lower Critical going low  | Asserted | Reading -2560 < Threshold 585 RPM
2e | 08/08/2015 | 07:29:38 | Fan FAN 2 | Lower Non-recoverable going low  | Asserted | Reading -2560 < Threshold 400 RPM
2f | 08/08/2015 | 08:53:06 | Temperature System Temp | Upper Non-critical going high | Asserted | Reading 110 > Threshold 75 degrees C
30 | 08/08/2015 | 08:53:06 | Temperature System Temp | Upper Critical going high | Asserted | Reading 110 > Threshold 77 degrees C
31 | 08/08/2015 | 08:53:06 | Temperature System Temp | Upper Non-recoverable going high | Asserted | Reading 110 > Threshold 79 degrees C
32 | 08/09/2015 | 00:58:08 | Fan FAN 5 | Lower Non-critical going low  | Asserted | Reading -2560 < Threshold 770 RPM
33 | 08/09/2015 | 00:58:08 | Fan FAN 5 | Lower Critical going low  | Asserted | Reading -2560 < Threshold 585 RPM
34 | 08/09/2015 | 00:58:08 | Fan FAN 5 | Lower Non-recoverable going low  | Asserted | Reading -2560 < Threshold 400 RPM
35 | 08/09/2015 | 06:27:06 | Temperature System Temp | Upper Non-recoverable going high | Deasserted | Reading 13 > Threshold 79 degrees C
36 | 08/09/2015 | 06:27:06 | Temperature System Temp | Upper Critical going high | Deasserted | Reading 13 > Threshold 77 degrees C
37 | 08/09/2015 | 06:27:06 | Temperature System Temp | Upper Non-critical going high | Deasserted | Reading 13 > Threshold 75 degrees C
38 | 08/10/2015 | 14:22:08 | Fan FAN 4 | Lower Non-critical going low  | Asserted | Reading -2560 < Threshold 770 RPM
39 | 08/10/2015 | 14:22:08 | Fan FAN 4 | Lower Critical going low  | Asserted | Reading -2560 < Threshold 585 RPM
3a | 08/10/2015 | 14:22:08 | Fan FAN 4 | Lower Non-recoverable going low  | Asserted | Reading -2560 < Threshold 400 RPM
3b | 08/13/2015 | 06:52:40 | Fan FAN 1 | Upper Non-critical going high | Asserted | Reading 33700 > Threshold 29260 RPM
3c | 08/13/2015 | 06:52:40 | Fan FAN 1 | Upper Critical going high | Asserted | Reading 33700 > Threshold 29815 RPM
3d | 08/13/2015 | 06:52:40 | Fan FAN 1 | Upper Non-recoverable going high | Asserted | Reading 33700 > Threshold 30370 RPM
3e | 08/13/2015 | 13:49:07 | Fan FAN 1 | Upper Non-recoverable going high | Deasserted | Reading 15200 > Threshold 30370 RPM
3f | 08/13/2015 | 13:49:07 | Fan FAN 1 | Upper Critical going high | Deasserted | Reading 15200 > Threshold 29815 RPM
40 | 08/13/2015 | 13:49:07 | Fan FAN 1 | Upper Non-critical going high | Deasserted | Reading 15200 > Threshold 29260 RPM
41 | 08/16/2015 | 21:35:35 | Fan FAN 3 | Lower Non-critical going low  | Asserted | Reading -2560 < Threshold 770 RPM
42 | 08/16/2015 | 21:35:35 | Fan FAN 3 | Lower Critical going low  | Asserted | Reading -2560 < Threshold 585 RPM
43 | 08/16/2015 | 21:35:35 | Fan FAN 3 | Lower Non-recoverable going low  | Asserted | Reading -2560 < Threshold 400 RPM
44 | 08/18/2015 | 22:34:50 | Fan FAN 1 | Lower Non-critical going low  | Asserted | Reading -2560 < Threshold 770 RPM
45 | 08/18/2015 | 22:34:50 | Fan FAN 1 | Lower Critical going low  | Asserted | Reading -2560 < Threshold 585 RPM
46 | 08/18/2015 | 22:34:50 | Fan FAN 1 | Lower Non-recoverable going low  | Asserted | Reading -2560 < Threshold 400 RPM
47 | 08/18/2015 | 22:35:21 | Fan FAN 1 | Lower Non-recoverable going low  | Deasserted | Reading 955 < Threshold 400 RPM
48 | 08/18/2015 | 22:35:21 | Fan FAN 1 | Lower Critical going low  | Deasserted | Reading 955 < Threshold 585 RPM
49 | 08/19/2015 | 17:37:53 | Fan FAN 1 | Lower Critical going low  | Asserted | Reading -2560 < Threshold 585 RPM
4a | 08/19/2015 | 17:37:53 | Fan FAN 1 | Lower Non-recoverable going low  | Asserted | Reading -2560 < Threshold 400 RPM
4b | 08/19/2015 | 17:38:52 | Fan FAN 1 | Lower Non-recoverable going low  | Deasserted | Reading 955 < Threshold 400 RPM
4c | 08/19/2015 | 17:38:52 | Fan FAN 1 | Lower Critical going low  | Deasserted | Reading 955 < Threshold 585 RPM
4d | 08/23/2015 | 07:22:53 | Fan FAN 1 | Lower Non-critical going low  | Deasserted | Reading 10390 < Threshold 770 RPM
4e | 08/24/2015 | 06:23:50 | Fan FAN 1 | Lower Non-critical going low  | Asserted | Reading -2560 < Threshold 770 RPM
4f | 08/24/2015 | 06:23:50 | Fan FAN 1 | Lower Critical going low  | Asserted | Reading -2560 < Threshold 585 RPM
50 | 08/24/2015 | 06:23:50 | Fan FAN 1 | Lower Non-recoverable going low  | Asserted | Reading -2560 < Threshold 400 RPM
51 | 08/24/2015 | 06:24:49 | Fan FAN 1 | Lower Non-recoverable going low  | Deasserted | Reading 955 < Threshold 400 RPM
52 | 08/24/2015 | 06:24:49 | Fan FAN 1 | Lower Critical going low  | Deasserted | Reading 955 < Threshold 585 RPM
53 | 08/24/2015 | 20:16:23 | Fan FAN 1 | Lower Critical going low  | Asserted | Reading -2560 < Threshold 585 RPM
54 | 08/24/2015 | 20:16:23 | Fan FAN 1 | Lower Non-recoverable going low  | Asserted | Reading -2560 < Threshold 400 RPM
55 | 08/24/2015 | 20:16:51 | Fan FAN 1 | Lower Non-recoverable going low  | Deasserted | Reading 955 < Threshold 400 RPM
56 | 08/24/2015 | 20:16:51 | Fan FAN 1 | Lower Critical going low  | Deasserted | Reading 955 < Threshold 585 RPM


Zusätzlich habe ich jetzt mal sensors nicht geladen. mal sehen ob es dann trotzdem zum Piepen kommt.

morlix
Community Developer
Community Developer
Posts: 175
Joined: July 18th, 2011, 10:08 am
Location: Stuttgart

Re: IPFire Sensors <-> Supermicro IPMI

Post by morlix » August 28th, 2015, 6:30 am

Kannst du von den Zeitstempeln her ein matching mit dem starten des piepen machen?

User avatar
tam
Posts: 42
Joined: December 6th, 2007, 8:10 am

Re: IPFire Sensors <-> Supermicro IPMI

Post by tam » August 28th, 2015, 7:02 am

Bin mir ziemlich sicher, dass die Uhrzeit nicht stimmt, und außerdem kriege ich nicht immer sofort mit, wenn es anfängt zu Piepen. Stand im Moment ist, dass ohne geladenes sensors bisher nix piept.

Aber ich traue diesen Sensoren eh nicht, zeigen sie doch z.B. als Minimumtemperatur der CPU -65°C an :o und die Lüfterdrehtzahl -2560 kllingt ja auch nicht irgendwie realitisch.

Post Reply