[f-nsp] foundry-nsp Digest, Vol 98, Issue 24
Wilbur Smith
wsmith at brocade.com
Thu Mar 31 12:44:46 EDT 2011
Harry,
I think you're ok. There are multiple temp sensors on each module, usually at the front where fresh air is drawn in, and at the rear when it's being exhausted. It's normal for the sensors in the rear to run hotter because there reading the increased temp after air flows over the heat exchangers. It's also normal for some slots to run a bit hotter than others. The type of card makes a large difference too...4x10G XFP is much hotter than the newer 8x10G SFP+.
I'd use Sensor #1 as your baseline sensor as I think that's the one measuring the air as it's drawn in. Also, are your fans running on HIGH, or cycling between Medium and High? Are you seeing any syslog messages warning about high temps?
Wilbur
-----Original Message-----
From: foundry-nsp-bounces at puck.nether.net [mailto:foundry-nsp-bounces at puck.nether.net] On Behalf Of foundry-nsp-request at puck.nether.net
Sent: Thursday, March 31, 2011 10:00 AM
To: foundry-nsp at puck.nether.net
Subject: foundry-nsp Digest, Vol 98, Issue 24
Send foundry-nsp mailing list submissions to
foundry-nsp at puck.nether.net
To subscribe or unsubscribe via the World Wide Web, visit
http://puck.nether.net/mailman/listinfo/foundry-nsp
or, via email, send a message with subject or body 'help' to
foundry-nsp-request at puck.nether.net
You can reach the person managing the list at
foundry-nsp-owner at puck.nether.net
When replying, please edit your Subject line so it is more specific
than "Re: Contents of foundry-nsp digest..."
Today's Topics:
1. XMR Temperature sensors (Harry Lewins)
----------------------------------------------------------------------
Message: 1
Date: Thu, 31 Mar 2011 12:20:09 +0100
From: "Harry Lewins" <Harry.Lewins at hso.uk.com>
To: <foundry-nsp at puck.nether.net>
Subject: [f-nsp] XMR Temperature sensors
Message-ID:
<CCD898B1733B354687D9689A1E8EDD5C014CD957 at wdh3-74757.hso.local>
Content-Type: text/plain; charset="us-ascii"
Hi guys,
Wondering if you could give me a bit of steer here... We have multiple
XMR-4s and 16s in our core, and we pretty varied temperatures on them,
despite them all being in temperature controlled data centres etc.
Within this, we also see massive differences on the individual sensors.
For example, an XMR-4:-
SLOT #: CARD TYPE: SENSOR # TEMPERATURE (C):
33 ACTIVE MG 1 35.500C
33 ACTIVE MG 2 48.500C
=======================================================
34 STANDBY MG 1 38.500C
=======================================================
1 LP 1 36.500C
1 LP 2 55.625C
2 LP 1 49.500C
2 LP 2 72.0C
SNM # FE # TEMPERATURE (C):
1 N/A 34.0C
2 N/A 35.5Ce
One sensor is over 72 degrees!
This XMR-16 varies from 35 to 59 degrees:-
SLOT #: CARD TYPE: SENSOR # TEMPERATURE (C):
33 ACTIVE MG 1 42.0C
33 ACTIVE MG 2 61.250C
=======================================================
34 STANDBY MG 1 36.500C
=======================================================
1 LP 1 47.500C
1 LP 2 46.500C
2 LP 1 39.0C
2 LP 2 43.500C
3 LP 1 43.0C
3 LP 2 45.0C
4 LP 1 40.500C
4 LP 2 48.500C
5 LP 1 44.500C
5 LP 2 48.500C
6 LP 1 40.0C
6 LP 2 42.250C
7 LP 1 45.500C
7 LP 2 46.500C
8 LP 1 37.500C
8 LP 2 38.500C
9 LP 1 43.500C
9 LP 2 59.0C
15 LP 1 49.0C
15 LP 2 51.250C
15 LP 3 UNUSED
16 LP 1 43.0C
16 LP 2 42.0C
16 LP 3 UNUSED
SNM # FE # TEMPERATURE (C):
1 N/A 42.0C
2 N/A 35.0C
3 N/A 38.5C
This XMR-4 varies from 29.5 to over 60 degrees...
SLOT #: CARD TYPE: SENSOR # TEMPERATURE (C):
33 ACTIVE MG 1 29.500C
33 ACTIVE MG 2 41.125C
=======================================================
1 LP 1 33.0C
1 LP 2 47.875C
2 LP 1 42.0C
2 LP 2 60.250C
SNM # FE # TEMPERATURE (C):
1 N/A 27.0C
2 N/A 29.5C
I could go on, but I guess you get the point! I'm just wondering if this
is normal - obviously I would expect some variation in the readings,
depending on where the sensor is in the chassis, how populated it is,
what cards are in it, how busy those cards are etc. But this seems very
high, and indeed temperatures over 70 degrees sound like a bad thing to
me...
Do you guys see similar temperatures? Initially we were monitoring all
these sensors with SNMP and freaking out every time one went over 60
degrees, but we've scaled that back a bit now. Do you guys just monitor
the cooler sensor? That is our current plan.
Does anyone know at what temperature we would start seeing a degraded
performance? I'm waiting on our supplier to give me some documentation
on this, but just wondered if you guys have real world experience.
Thanks for your help,
Harry
_____________________________________________________________________
This e-mail and all attachments have been scanned by the hSo virus scanning service powered by Webroot and no known viruses were detected.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/foundry-nsp/attachments/20110331/e2831ca5/attachment-0001.html>
------------------------------
_______________________________________________
foundry-nsp mailing list
foundry-nsp at puck.nether.net
http://puck.nether.net/mailman/listinfo/foundry-nsp
End of foundry-nsp Digest, Vol 98, Issue 24
*******************************************
More information about the foundry-nsp
mailing list