Dell N1548 - FAILURE event : fan

Bonjour,

Sur deux nouveaux N1548, j'ai des erreur de ventilateurs qui apparaissent de temps à autres :

<187> Oct 12 23:44:46 SWMAQ01-1 BOXSERV[boxs Req]: boxs.c(1171) 16421 %% Unit 1 fan 2 FAILURE event (4) occurred.

<187> Oct 13 02:06:17 SWMAQ01-1 BOXSERV[boxs Req]: boxs.c(1171) 22519 %% Unit 1 fan 2 FAILURE event (4) occurred.

<187> Oct 13 02:53:48 SWMAQ01-1 BOXSERV[boxs Req]: boxs.c(1171) 24530 %% Unit 1 fan 1 FAILURE event (4) occurred.

<187> Oct 13 03:09:51 SWMAQ01-1 BOXSERV[boxs Req]: boxs.c(1171) 25256 %% Unit 1 fan 2 FAILURE event (4) occurred.

etc...

L'état du switch me semble pourtant OK :

> show system

System Description: Dell Networking Switch
System Up Time: 0 days, 15h:25m:11s
System Contact:
System Name:
System Location:
Burned In MAC Address: 1418.77E8.F536
System Object ID: 1.3.6.1.4.1.674.10895.3065
System Model ID: N1548
Machine Type: Dell Networking N1548

System Thermal Conditions:

Unit Temperature State
(Celsius)
---- ----------- -----------
1 38 Good

Temperature Sensors:

Unit Description Temperature
(Celsius)
---- ------------------ -----------
1 MAC 38
1 PHY 34

Fans:

Unit Description Status
---- ----------- ------
1 Fan-1 OK
1 Fan-2 OK

Power Supplies:

Unit Description Status Average Current Since
Power Power Date/Time
(Watts) (Watts)
---- ----------- ------------ ---------- -------- -------------------
1 System OK 25.5 25.3
1 Main OK N/A N/A 10/12/2017 15:43:27
1 RPS No Power

USB Port Power Status:
----------------------
Device Not Present

> show version

Machine Description............... Dell Networking Switch
System Model ID................... N1548
Machine Type...................... Dell Networking N1548
Serial Number..................... CN0V143P282987120035A02
Manufacturer...................... 0xbc00
Burned In MAC Address............. 1418.77E8.F536
System Object ID.................. 1.3.6.1.4.1.674.10895.3065
SOC Version....................... BCM56150_A0
HW Version........................ 2
CPLD Version...................... 16

unit active backup current-active next-active
---- ----------- ----------- -------------- --------------
1 6.3.3.8 6.2.6.6 6.3.3.8 6.3.3.8

J'ai redémarré plusieurs fois les switchs et les erreurs apparaissent encore :

<187> Oct 13 21:24:24 SWMAQ02-1 BOXSERV[boxs Req]: boxs.c(1171) 685 %% Unit 1 fan 2 FAILURE event (4) occurred.
<187> Oct 13 23:37:23 SWMAQ02-1 BOXSERV[boxs Req]: boxs.c(1171) 694 %% Unit 1 fan 2 FAILURE event (4) occurred.
<187> Oct 14 00:15:50 SWMAQ02-1 BOXSERV[boxs Req]: boxs.c(1171) 700 %% Unit 1 fan 2 FAILURE event (4) occurred.
<187> Oct 14 05:27:44 SWMAQ02-1 BOXSERV[boxs Req]: boxs.c(1171) 707 %% Unit 1 fan 2 FAILURE event (4) occurred.
<187> Oct 14 05:41:37 SWMAQ02-1 BOXSERV[boxs Req]: boxs.c(1171) 714 %% Unit 1 fan 2 FAILURE event (4) occurred.
<187> Oct 14 06:59:36 SWMAQ02-1 BOXSERV[boxs Req]: boxs.c(1171) 733 %% Unit 1 fan 2 FAILURE event (4) occurred.
<187> Oct 14 08:27:43 SWMAQ02-1 BOXSERV[boxs Req]: boxs.c(1171) 741 %% Unit 1 fan 2 FAILURE event (4) occurred.
<187> Oct 14 09:17:23 SWMAQ02-1 BOXSERV[boxs Req]: boxs.c(1171) 749 %% Unit 1 fan 1 FAILURE event (4) occurred.
<187> Oct 14 10:18:49 SWMAQ02-1 BOXSERV[boxs Req]: boxs.c(1171) 756 %% Unit 1 fan 1 FAILURE event (4) occurred.

J'ai également constaté que les vitesses des ventilateurs ne sont pas régulières :

Pourtant, la température des capteurs est constante (38 °C), la charge cpu est à 18% :

SWMAQ02#show process cpu

Memory Utilization Report

status KBytes
------ ----------
free 614016
alloc 400828

CPU Utilization:

PID Name 5 Secs 60 Secs 300 Secs
---------- ------------------- -------- -------- --------
3 (ksoftirqd/0) 0.00% 0.07% 0.06%
1057 (max3421_spi_thr) 0.19% 0.05% 0.03%
1265 (procmgr) 0.38% 0.22% 0.19%
1382 osapiTimer 0.00% 0.02% 0.02%
1384 bcmINTR 0.38% 0.62% 0.44%
1385 socdmadesc.0 0.00% 0.07% 0.15%
1386 socdmadesc.1 0.19% 0.19% 0.16%
1387 bcmMEM_SCAN.0 0.38% 0.23% 0.23%
1389 bcmL2X.0 3.27% 3.25% 3.45%
1390 bcmCNTR.0 0.19% 0.46% 0.58%
1393 bcmMEM_SCAN.1 0.57% 0.30% 0.29%
1395 bcmL2X.1 3.27% 3.55% 3.64%
1396 bcmCNTR.1 0.96% 0.68% 0.66%
1397 envMonitorTask 0.00% 0.08% 0.08%
1401 bcmRX 0.77% 0.39% 0.35%
1414 bcmLINK.0 2.50% 2.34% 2.27%
1415 bcmLINK.1 2.31% 2.19% 2.24%
1416 cpuUtilMonitorTask 0.38% 0.33% 0.32%
1424 tL7Timer0 0.00% 0.02% 0.00%
1429 syncdbAgentTask 0.00% 0.01% 0.00%
1432 tap_monitor_task 0.19% 0.17% 0.14%
1445 emWeb 0.38% 0.09% 0.05%
1447 dtlTask 0.38% 0.27% 0.29%
1451 hapiRxTask 0.00% 0.15% 0.17%
1471 SNMPTask 0.19% 0.15% 0.24%
1494 snoopTask 0.00% 0.05% 0.08%
1524 ipMapForwardingTask 0.19% 0.29% 0.26%
1530 openrTask 1.15% 1.12% 1.13%
1555 lldpTask 0.00% 0.02% 0.02%
1566 RMONTask 0.19% 0.10% 0.10%
1580 StatsAppTask 0.38% 0.37% 0.39%
------------------------------ -------- -------- --------
Total CPU Utilization 18.89% 18.00% 18.21%

La configuration de ces châssis est très simple. Ils sont actuellement en préparation pour une future mise en Prod et très peu de trafic sur le fond de panier. 

Nous possédons d'autre Dell N séries 2000 et 4000. Seules les séries 1500 me posent actuellement ce problème.

La version du firmware est à priori la dernière : 6.3.3.8

Auriez-vous une idée ?

Merci d'avance.

Cordialement,