DL 380 G9 High fan speed

[Mano Hernandez]

We just purchased 7 G9's and installed windows 2012. However one of them after about 20 minutes, the fans kick in really loud and just roar. They do not stop. a restart will fix it, but that's not the solution. It seems to me the issue started when I ran all of the Windows updates on this one machine. It's not in production yet, but it will be.

[Jimmy Vance]

Have you looked at the IML or iLO event logs?

[Webdoggie]

Was there a resolution on this item? I am seeing the same thing on my DL380 G9. High Fan speed with no entries in the logs that note any state changes.

Rather than reboot, connecting to the ILO and restarting it relieves the condition.

ILO is 2.40. System ROM is P89 v2.20 (6/2/2016). All system info is showing green checkmarks, even when the fans are running a full blast.

Power Management Controller is v1.0.9

Any resolution would be helpful. Have to restart the ILO about 1 or 2 times per day about every 4th or 5th day.

Thanks

Ron Eubank
PYRAMID

[Webdoggie]

Ach, nevermind....

http://h20565.www2.hpe.com/hpsc/doc/public/display?sp4ts.oid=7271241&docLocale=en_US&docId=emr_na-c05098585

The above link contains the solution. If web pages change, look for Document ID c05098585 Version 3,  

Advisory: (Revision) Certain ProLiant Gen8 and Gen9 Servers - Fans May Run at a High Speed After Updating The Adapter Firmware with Version 4.0.1.17 of the HPE Broadcom NX1 Online Firmware Upgrade Utility for Windows

~re

[ziasiddiqui]

I've faced the same issue when I purchased this server. It has something to do with the firmware upgrade as initially, I was not able to go to provisioning option. Get the latest firmware upgrade (i.e. 2.5 or 2.6) from enterprise support website and install it.

This will solve the issue.

[jonahgroup]

It doesn't solve the problem for me.

And unfortunately this server recently fell out of warranty status, so HP won't even fart in my general direction.

No errors in the health logs.  I've tried updating everything with SPP 2018.3, SPP 2017.10 and 2017.7, and the problem still won't go away.  It's been like this for almost a week now.

And now one of the hard drives has failed.  Don't know if the constant high fans rattled its connections loose or something, but now I can't even experiment further until the array rebuilds.  Don't know if that also causes the fans to go crazy, but until the array is rebuilt I can't play anymore firmware flash merry go round.

[Jimmy Vance]

---

@jonahgroup wrote:

It doesn't solve the problem for me.

And unfortunately this server recently fell out of warranty status, so HP won't even fart in my general direction.

No errors in the health logs.  I've tried updating everything with SPP 2018.3, SPP 2017.10 and 2017.7, and the problem still won't go away.  It's been like this for almost a week now.

And now one of the hard drives has failed.  Don't know if the constant high fans rattled its connections loose or something, but now I can't even experiment further until the array rebuilds.  Don't know if that also causes the fans to go crazy, but until the array is rebuilt I can't play anymore firmware flash merry go round.

---

What does iLO report the fan speeds at? Have you looked at the system tempatures in iLO to see if something is running warm? Has the ambient room tempature increased?  Verify in RBSU the thermal settings are not set to max or increased cooling as those option will spin the fans faster.  You checked both iLO and the IML logs?

.  

[jonahgroup]

We were repurposing the machine, and in the process I reconfigured the array from JBOD to RAID10, and I applied SPP2018.3.  The problems immediately started occurring after that.  There were zero problems prior.

The environment itself (server room) is identical to what it was before.  I checked the bios system health report and everything showed up as Healthy.  I haven't had time to dig further than that.

[jonahgroup]

I've revert all firmwares on the system to whatever is provided by SPP 2017.4.  Everything appears to have calmed down now.  I will have to wait and see if it stays that way.

But we lost *4* hard drives in 2 days because of this.

And to answer your questions, everything had been reset to defaults, so the power had been configured for "Balanced".  Nothing had changed environment-wise, and the fans would kick in almost immediately after boot, before the primary OS had even started.

I didn't check what iLO was reporting before I reverted everything, but at this point all sensors are nominal.

I have to say I am thoroughly disgusted with HP.  Between poor quality VMWare drivers that purple screen the entire host

I very much regret buying HP.

[jonahgroup]

The fans have kicked on high again.  iLO shows them running at 73%.

All temperatures are reported within good range.

Of the highest values I'm seeing:

08-HD Max is at 47C
02-CPU 1 is at 40C
19-PS 1 Internal is at 40C
20-PS 2 Internal is at 40C
27-HD Controller is at 48C
29-LOM is at 40C

All other sensors are in the low 20s or less.  Everything is well below the "Caution" limit.

Room temperature is only 18C.

The only errors in the iLO event log is that it can't reach the NTP server.

These values don't seem generally that high to me.

Thoughts?