Unsolved
This post is more than 5 years old
38 Posts
0
98618
M1000e fans and the M610x...what's happening?
Has anyone had a problem with the M610x causing all the fans in the chassis to ramp up to 14-15K RPM and stay there?
We have M1000e with:
M610x in slots (1&9) and (2&10)
M600 in slots 3,4 5,6,7,8,11 & 12
M610 in slots 13 & 14
Chassis power capped at 5500W. 6 PSUs give 2360W each.
Without the M610x servers on, all the fans run between 4.7K and 6.1K RPM...temps are low and nowwhere near the popwer cap I've set.
Switch an M610x and it takes off...all fans immediately up over 14K and they stay there.
I've just spent 8 hours overtime updating every firmware I could find and still no joy.
And by the way, it's a show stopper. The extra power drawn by the chassis when all fans hit the roof takes us periolously close to our allowance in that rack. Are Dell aware of this? Is there a firmware fix on the way? Anything to give me hope?
I have a call raised with ProSupport (UK) 826469471 and have also found this poor soul with a similar issue: http://www.delltechcenter.com/thread/4317201/M1000e+fan+control
Is there an issue with the BIOS/CPLD of the M610x that needs looking at maybe?
Regards,
Neil
We have M1000e with:
M610x in slots (1&9) and (2&10)
M600 in slots 3,4 5,6,7,8,11 & 12
M610 in slots 13 & 14
Chassis power capped at 5500W. 6 PSUs give 2360W each.
Without the M610x servers on, all the fans run between 4.7K and 6.1K RPM...temps are low and nowwhere near the popwer cap I've set.
Switch an M610x and it takes off...all fans immediately up over 14K and they stay there.
I've just spent 8 hours overtime updating every firmware I could find and still no joy.
And by the way, it's a show stopper. The extra power drawn by the chassis when all fans hit the roof takes us periolously close to our allowance in that rack. Are Dell aware of this? Is there a firmware fix on the way? Anything to give me hope?
I have a call raised with ProSupport (UK) 826469471 and have also found this poor soul with a similar issue: http://www.delltechcenter.com/thread/4317201/M1000e+fan+control
Is there an issue with the BIOS/CPLD of the M610x that needs looking at maybe?
Regards,
Neil
grahammeadows
3 Posts
0
December 8th, 2010 01:00
nrawlinson1000
38 Posts
0
December 8th, 2010 10:00
NumaoMasayuki
9 Posts
0
December 9th, 2010 01:00
The processors in M610x are Xeon L5640 * 2 with 96GB memory.
BIOS version is 2.1.16.
I already informed customer support of DELL Japan, and they replaced momory unit because CMC showed "Mem ECC Warning" but after a couple of hours, the same warning appears and the fans are keeping the maximum speed.
I do not know how to "open a support case" other than just claiming to DELL support.
Kong Yang
180 Posts
0
December 9th, 2010 07:00
If you called Dell customer support and opened a case, they would assign you a support #. Stay tuned as PG and IPS are finalizing root-cause analysis.
KongY@Dell
KongY@Dell
Kong Yang
180 Posts
0
December 9th, 2010 07:00
Sorry to hear about more of your issues. Have you opened a support case for the M600s? I will ping PG and IPS again.
KongY@Dell
nrawlinson1000
38 Posts
0
December 9th, 2010 07:00
More problems though - after we've upgraded all the firmwares on the chassis and our existing M600, we've hit other problems worse than the first!
Our M600s stop communication with the chassis - and then come back with no intervention - not always at the same time. iDRACs stop responding and then come back to life with no intervention. The only symptom we see is in Windows System eventlog IPMIDRV 1004 warnings. But if we're unlucky and a couple of servers in a cluster get struck at the same time, we lose Exchange or our Hyper-V cluster. Not good. Not good at all.
I've raised a separate thread here... http://www.delltechcenter.com/thread/4378686/IPMIDRV+1004+errors+on+Windows+2008+R2+-+M600+blades
grahammeadows
3 Posts
0
December 9th, 2010 08:00
However in my current situation - fans running at full speed - there are no error lights on the blades and no indication of problems on any of them. Unfortunately they are all part of a critical system so it is going to be a while before I am allowed to power them down.
Also the call I logged earlier - the advice was to update the BIOS and firmware on the chassis and blades. Following your comments nrawlinson I might hold off doing that!!
darmstrong.navi
7 Posts
0
December 9th, 2010 08:00
nrawlinson1000
38 Posts
0
December 9th, 2010 08:00
nrawlinson1000
38 Posts
0
December 9th, 2010 09:00
http://www.delltechcenter.com/thread/4378686/IPMIDRV+1004+errors+on+Windows+2008+R2+-+M600+blades
I have been impressed with the way this forum works, and the fact that Dell are so responsive.
nrawlinson1000
38 Posts
0
December 10th, 2010 08:00
From looking at the server we sent off to Dell, they've discovered a problem between the iDRAC and the "low-power" Intel CPUs in there. Ours are the L5640 2.26Ghz. A new iDRAC firmware is in testing, assuming all goes well this should be available some time next week.
Does this tie in with anyone else's problem - especially Numao perhaps?
NumaoMasayuki
9 Posts
0
December 14th, 2010 18:00
The difference from nrawlinson's case is that once we remove the GPU card from M610x, the fans calm down to the normal speed.
Any good news?
nrawlinson1000
38 Posts
0
December 14th, 2010 23:00
nrawlinson1000
38 Posts
0
December 17th, 2010 03:00
nrawlinson1000
38 Posts
0
December 22nd, 2010 13:00
Neil