AnonP709da6

updated

10 days ago

A

AnonP709da6

1 Rookie

•

3 Posts

•

24 Points

0

94

November 4th, 2026 08:49

OME has been unusable since 4.7 for us

Since 4.7 (and continuing with 4.8) I struggle to update the firmware on my environment.  I have a mixture of PowerEdge m1000E chassis, M640, MX7000 chassis and MX750 and MX760 servers.

Pushing out updates fail for various reasons.

 

  • Verifying if the device Service Tag is valid. The device Service Tag is valid. Starting communication with the device. Exception while executing action: GET_LOCKDOWN_STATUSCannot invoke "java.lang.Integer.intValue()" because "httpResponseCode" is null Task Failed. Completed With Errors.
  • Verifying if the device Service Tag is valid. The device Service Tag is valid. Starting communication with the device. Task Failed. Completed With Errors. 
  • Verifying if the device Service Tag is valid. The device Service Tag is valid. Unable to perform the management task on this device because the device is not onboarded. Task Failed. Completed With Errors. 

I'm three days in of trying to update iDRAC firmware. Usually this would be one job and complete within a morning. Instead I am having to keep trying the job in smaller batches. Servers will fail, and then if I retry, randomly work. I've ran multiple discovery and inventory jobs (which also are scheduled over night).

I never had these issue with 4.6 but since 4.7 and now 4.8 it seems patching would be quicker if I did it manually server by server.

 

Is nobody else experiencing issues?

  • DELL-Bridjet V

    Moderator

    •

    118 Posts

    •

    338 Points

    0

    0

    Posted November 6th, 2026 04:12

    Hi @AnonP709da6,

    The errors you are encountering—specifically the GET_LOCKDOWN_STATUS null pointer exception, intermittent connection drops, and the device is not onboarded error—point to communication timeouts, session exhaustion, or Redfish API/credential cache desynchronization between OpenManage Enterprise (OME) and the target out-of-band controllers (iDRAC9 and CMC).

    OME 4.x introduced architectural changes (moving to containerized microservices platform on Photon OS), and versions 4.7/4.8 enforce stricter Redfish API timeouts, modern cipher suites, and concurrent connection handling that frequently expose underlying session or resource constraints.

    Here are the primary causes and how to resolve them:

     

    1. Appliance Resource Sizing (OME 4.x VM)

    OME 4.x requires significantly more resources than older releases, especially when managing modular infrastructure (M1000e, MX7000) alongside 14G, 15G, and 16G blades:

    • Check VM Sizing: Ensure your OME appliance has at least 8 vCPUs and 32 GB of RAM (or minimum 16 GB for smaller fleets).
    • When the appliance runs low on memory or CPU cycles, the internal Redfish proxy drops calls mid-flight, returning a null HTTP response code (Cannot invoke "java.lang.Integer.intValue()" because "httpResponseCode" is null).
    • Action: Shut down the OME VM, verify/increase RAM and vCPU allocations according to Dell's sizing guidelines, and restart the appliance.

     2. Clear iDRAC and CMC Web Server / Session Exhaustion

    The error httpResponseCode is null means OME initiated a TCP connection, but the iDRAC/CMC dropped or reset the socket before sending an HTTP response. This commonly happens when:

    • An iDRAC's internal web server is out of memory or has hung active sessions.
    • Multiple parallel inventory, discovery, and update jobs have exhausted the maximum allowed concurrent sessions on the iDRAC.

    Action:

    • Perform a soft reset of the iDRAC on a few failing servers:

    bash

    racadm racreset

    (This only resets the management controller; it does not affect running operating systems or workloads).

    • For the M1000e, ensure the active Chassis Management Controller (CMC) is healthy and not reporting high memory utilization.

     3. Address the "Device is not onboarded" Error

    This error occurs when the security token, certificate handshake, or Redfish session trust between OME and the device is invalidated or out of sync.

    • Re-run Discovery with the "Manage" Option:
      1. Go to Devices > select the problematic server(s).
      2. Select More Actions > Discover.
      3. Re-enter the device credentials and ensure Manage (not just "Monitor") is selected so the onboarding certificate is re-exchanged.
    • Certificate / Cipher Check:
      • OME 4.7/4.8 restricts deprecated SSL/TLS ciphers. If an older modular appliances like M1000e CMCs or M640 iDRACs are on significantly older firmware, the TLS handshake may intermittently fail.
      • Ensure M640 iDRACs and M1000e CMCs are updated to their latest available baseline firmware to support modern TLS 1.2/1.3 ciphers.

     4. Reduce Parallelism and Stagger Schedules

    Because modular chassis (MX7000 with OME-Modular and M1000e with CMC) share internal backplane management paths to individual blades, pushing updates to multiple blades in the same chassis simultaneously can saturate the internal management switch.

    • Split by Chassis: Rather than pushing firmware across the entire fleet, target one chassis at a time.
    • Disable Overlapping Tasks: Ensure your scheduled discovery/inventory tasks are not overlapping with your deployment windows. If an automated inventory job kicks off while a deployment job is interrogating lockdown status, the jobs conflict and fail.

     5. Clear OME Appliance Session Cache

    If the internal API gateway within the OME appliance is holding stale connection states:

    1. Log in to the OME appliance Text User Interface (TUI) via the VM console.
    2. Select Restart Services (or reboot the entire appliance).
    3. Allow 10–15 minutes for all microservice pods to initialize fully before launching deployment tasks.

    Regards,

    DELL-Bridjet V

    Social Media and Communities Professional

    Dell Technologies | Enterprise Support Services

    #IWork4Dell

    Was your query answered? Please click Thumbs Up on the post that helped to Mark it as Accepted Answer!

  • AnonP709da6

    1 Rookie

    •

    3 Posts

    •

    24 Points

    2

    0

    Posted November 6th, 2026 08:04

    @DELL-Bridjet V​ thanks for the in-depth, detailed reply!

     

    As a last resort, yesterday, I deleted all devices from my inventory and then ran a rediscover job. After that, I didn't see any more failures. 

    When OME v4 was released we decided to set it up as new rather than trying to upgrade from v3. So the whole setup is not really that old. 

    Our appliance is configured with 8CPUs and 64GB of RAM. We only have 174 devices, according to OME, so this should be more than enough to manage this environment.  

    I did try doing servers in batches. It would even fail if I just selected one server. 

    Ultimately, deleting all the devices and rediscovering them seems to have helped.

    I will see what happens next time we need to push out updates and consider your answers. Again, thanks for the responce.

  • DELL-Bridjet V

    Moderator

    •

    118 Posts

    •

    338 Points

    1

    0

    Posted November 6th, 2026 14:18

    Hi @AnonP709da6,

    You are very welcome. I'm pleased to hear that the issue has been resolved.

    Regards,

    DELL-Bridjet V

    Social Media and Communities Professional

    Dell Technologies | Enterprise Support Services

    #IWork4Dell

    Was your query answered? Please click Thumbs Up on the post that helped to Mark it as Accepted Answer!