UNSOLVED

budjbsofd

updated

1 year ago

B

budjbsofd

1 Rookie

•

7 Posts

0

2992

December 13th, 2020 11:00

Dell T340 with Windows Server 2019 stuck at low CPU speed.

I have a T340 with a xeon e-2236 that won't seem to spin up past 0.78 ghz in Windows Server 2019. I have the profile in the bios set to performance and the windows power profile set the same.

I have put an artificial load on the system and gone though various software installations, but the system feels sluggish and the CPU speed never increases. I'm not sure where else to try adjusting settings to get my full speed (3.4 ghz). Any help would be appreciated!

  • DELL-Young E

    Moderator

    •

    5421 Posts

    •

    37 Points

    2002

    0

    Posted December 13th, 2020 18:00

    Hi can you change your BIOS profile?

    https://dell.to/3mi4znk

     

    Social Media and Communities Professional

    Dell Technologies | Enterprise Support Services

    #IWork4Dell

    Did I answer your query? Please click on ‘Mark as Accepted Answer’. ‘Thumbs up’ the posts you like!

  • DELL-Stefan R

    Moderator

    •

    790 Posts

    1985

    0

    Posted December 14th, 2020 06:00

    Hey there,

     

    yes, it looks like the CPU is the issue here, but still, there is a chance that the sensor on the mainboard is defective and is reporting wrong values (very vague I know, but still possible).

    If you can get your hands on a replacement CPU go and test it - this would be the best way to figure out if the CPU is defective.

    You might check it with the Intel CPU diag tool:

    https://downloadcenter.intel.com/download/19792/

     

    Maybe this shows different results.

     

    Best regards

    Stefan

     

  • budjbsofd

    1 Rookie

    •

    7 Posts

    1989

    0

    Posted December 14th, 2020 06:00

    I have changed the profile from performance to performance per watt and back with the same result.

    I was doing a little digging into why the CPU might be throttling and started looking at CPU temps and tried out Intel XTU. XTU reported that the CPU was definitely thermal throttling, which is odd because the CPU temps are typically locked in at 29C, even with full load (albeit, full load at around 800mhz).

    I next tried ThrottleStop, which gives you some more insights into the current state of the CPU and the ability to suppress various throttle signals. Using that, I was able to show that the CPU is indeed reporting BD_PROCHOT, and the CPU is indeed thermal throttling. When I suppressed the PROCHOT signal, the CPU ramps up just like it should to 3.41ghz, and even up to turbo speeds. Un-suppressing the PROCHOT signal throttles it right back down to around 800mhz.

    Edit: here's a link to a screenshot showing the ThrottleStop status https://photos.app.goo.gl/7Nuvykbvhz8HWKTVA

    I'm thinking this CPU may actually be defective. Do you agree?

  • budjbsofd

    1 Rookie

    •

    7 Posts

    1984

    0

    Posted December 14th, 2020 07:00

    This server is very much still under warranty. Unfortunately, I do not have a spare CPU laying around to swap in and test (we're a very small office). What's the best way to engage Dell over this issue?

    Also, a second question that may be tangentially related. I just put an H730p into this server. It was originally configured with software raid, but the system was nearly unusable (likely because of this actual CPU problem I just found). The new raid card works great, and even in its degraded state the system does respond faster since the CPU doesn't need to also handle RAID concerns. My question here is this: the server was configured with a 350W non-hot-swappable PSU. I am unable to find any specs about power requirements for the H730p. Should that 350W PSU be enough to handle this system AND the new raid card?

  • DELL-Marco B

    Moderator

    •

    3980 Posts

    1975

    0

    Posted December 14th, 2020 07:00

    Hello, 

    where the server is located? If the server is under warranty you can contact our technical support by phone, or we can also open a service call to resolve the issue.

    For the other question, unfortunately we don't have details about PERC watt consumption but I believe that 350W PSU is enough to manage the card, no warning are written in the server documentation about increase PSU power.

    Please let us know if you want to proceed with phone support or our community support for the service call.

     

    Thanks

    Marco

    DELL- Marco B

    Social Media and Communities Professional

    Dell Technologies | Enterprise Support Services

    #IWork4Dell

    Did I answer your query? Please click on ‘Mark as Accepted Answer’. ‘Thumbs up’ the posts you like!

  • budjbsofd

    1 Rookie

    •

    7 Posts

    1962

    0

    Posted December 14th, 2020 12:00

    I spoke with tech support and a tech is coming by tomorrow to swap the mainboard and CPU. I hope this fixes the problem, since if it doesn't I wonder if this is a firmware issue. I'll update the post with the results of the service tomorrow. Thanks for everyone's help so far!

  • 1876

    0

    Posted February 22nd, 2021 12:00

    How did things go?

  • HCR2025

    1 Rookie

    •

    5 Posts

    •

    2 Points

    0

    0

    Posted August 27th, 2025 15:36

    I am having the same issue as well, cannot update bios nor the lifecycle controller. Dell is coming to replace the motherboard today, will update this thread when i know something as well, in case others encounter this issue. I drained the flea power and got the core to come back up to its base clock and turbo, but it was at .78 for a while

  • HCR2025

    1 Rookie

    •

    5 Posts

    •

    2 Points

    0

    0

    Posted August 27th, 2025 15:40

    Update: IPMI in my case is banged out in device manager. 

  • HCR2025

    1 Rookie

    •

    5 Posts

    •

    2 Points

    0

    0

    Posted August 27th, 2025 15:49

    This is what the server said when it started up, the Lifecycle was gone for a while mind you, related \ unrelated to the issue at hand, however, again could not update Lifecycle nor Bios at the time of this writing.

    The server in fact hung here until i powered it off, this is the point the core came back to base clock, after the flea power drain, assuming by a fluke

    (edited)