Hi, a few months earlier, there are errors on both SP A and B bbu in Dell EMC VNXe3200 which also cause the DPE faulty errors to be displayed. Recently I changed the SP A and B bbu, and in the enclosures, it can be seen that the errors for both batteries are gone.
However the DPE faulty error is still there even though there are no more other subcomponents failed/faulty. Is there anything I can do? The software version is 3.1.12.10186894.
Thank you for reaching out and for providing the detailed information and screenshot.
It sounds like you have successfully replaced both the SP A and SP B Battery Backup Units (BBUs), and the battery-level faults have cleared. However, the parent-level DPE (Disk Processor Enclosure) fault can sometimes become a stale or persistent alert that does not automatically clear even after the underlying subcomponent issues are resolved. This is a known behavior on VNXe systems.
Here are the recommended steps to try to clear this persistent DPE fault:
1. Verify No Other Subcomponent Faults Remain
In Unisphere, navigate to System > System View > Enclosures and expand all subcomponents (SPs, Power Supplies, Fans, Disks, I/O Modules, Link Control Cards, etc.) to confirm that no other component is showing a degraded or faulted state.
Also check System > Alerts & Events for any active or unacknowledged alerts that may still be contributing to the DPE fault.
2. Acknowledge and Clear Any Outstanding Alerts
Go to System > Alerts and review all active alerts.
If there are any alerts related to the old BBU faults that were not acknowledged, select them and acknowledge/clear them.
3. Perform a Rolling Reboot of Both Storage Processors
A rolling reboot can often clear stale fault states at the DPE level:
Step 1: In Unisphere, navigate to Settings > More Configuration > Manage Storage Processors.
Step 2: Reboot SP A first. Wait for it to come fully back online and confirm it shows as Operating Normally.
Step 3: Then reboot SP B. Wait for it to come fully back online as well.
Important: Perform this as a rolling reboot (one SP at a time), never reboot both simultaneously. During a single SP reboot, the other SP will handle I/O, so there should be no data unavailability, though a brief performance impact is expected.
4. Check the DPE Status After Reboot
After both SPs have been rebooted and are operating normally, check whether the DPE fault has cleared. In many cases, the rolling reboot will resolve the stale fault condition.
I already performed the rolling reboot on both storage processors, however the DPE fault error still couldn't be cleared. If it is a stale alert, is it safe to ignore or there is any other way to clear the fault error?
Hi nafiz1102, I understand how frustrating it can be to see a persistent DPE fault even after the batteries that triggered it have been replaced and the storage processors have been rebooted. The situation you’re describing usually falls into one of three categories: the alert is truly stale, the DPE module still has an internal condition that hasn’t been cleared, or the firmware on the DPE or the SPs is out of sync with the hardware change and therefore continues to report the fault.
The first thing to confirm is that the DPE itself is reporting a healthy state. If you log into the SPs (either through the Unisphere shell or via SSH) and run the DPE status command, you should see a line indicating that the power‑supply and battery health are OK. On a VNXe the typical syntax is svc_dpe_status (or, on newer releases, svc_dpe show). If that command returns anything other than “OK” you’ll want to note the exact error text because it can give you a clue about whether the module still thinks a battery is missing, whether the charger is reporting a fault, or whether the DPE firmware needs to be refreshed.
Assuming the DPE status itself is clean, the next step is to clear the alert from the management database. In Unisphere you can usually do this by navigating to the Alerts view, selecting the DPE fault entry, and choosing “Clear.” That action simply removes the symptom from the user interface; it does not reset the hardware, which is why it sometimes reappears if the underlying condition isn’t truly resolved. If the UI does not allow you to clear it, you can issue the same operation from the command line with the alert‑clear utility: svctasks clearalert –id <alert‑id> (replace <alert‑id> with the numeric identifier shown in the alert details). Running that command after the DPE status shows healthy will typically force the management software to forget the old event.
When an alert still refuses to clear, a useful trick on the VNXe platform has been to force the SPs to re‑enumerate the DPE hardware. A rolling reboot is the first thing most people try, but sometimes the DPE firmware cache remains in memory. Executing a DPE service restart (for example, svc_restart_service restart DPE) forces the SP to reload the DPE driver and re‑read the hardware registers. After that service restart, give the array a minute or two to settle, verify the DPE status again, and then attempt to clear the alert once more. If the service restart command is not available on your version, the equivalent operation is to stop and start the “rsvd‐sp” service on each SP, which also triggers a hardware inventory refresh.
Another angle that occasionally shows up with VNXe 3.1.x is a mismatch between the battery firmware version and the DPE firmware version. The VNXe release notes for 3.1.12 mention that certain battery replacements required a DPE firmware update to prevent stale fault flags. If you have not yet applied the most recent DPE hot‑fix for your code level, it is worth checking the Dell EMC support site (or the firmware download portal) for a “DPE 3.1.x” update and applying it in the usual maintenance window. After the update, run the DPE status command again and clear the alert.
If you have verified that the DPE status is healthy, you have run a DPE service restart, you have applied the latest DPE firmware, and the alert still refuses to clear, the remaining option is to acknowledge that the alert is stale and suppress it using the alert‑suppress feature. This does not fix the underlying condition (because there isn’t one) but it prevents the alert from flooding the UI. The syntax for suppression is similar to the clear command, but you add the “–suppress” flag: svctasks clearalert –id <alert‑id> –suppress. Most administrators only use this as a last resort after they have documented that the hardware is healthy and the alert has been confirmed as a false positive.
In summary, I would recommend the following practical sequence: first, verify DPE health with svc_dpe_status; second, issue a DPE service restart (svc_restart_service restart DPE); third, try to clear the alert from the UI or with svctasks clearalert; fourth, confirm you are running the latest DPE firmware for 3.1.12; and finally, if the alert still persists, consider suppressing it. If after those steps the fault still appears, it usually indicates a deeper firmware inconsistency that would require a support case to get a guided firmware rollback or a more invasive hardware diagnostic. Even without an active support contract, you can still open a basic “hardware advisory” request to obtain the needed firmware packages or a clear‑alert script from Dell EMC. I hope this helps you get the DPE fault out of the way so you can move forward with confidence in your VNXe3200.
Hi Josh, thanks for the reply, but I couldn't run the command to check the DPE status, even with the newer release, and it is not even listed under the svc_. So what I can check instead is only using the svc_diag commands for both spA and spB, and both returns no error. Not sure whether those information are enough to claim that there are no faults in the VNXe3200.
As for the second option, I couldn't use the commands as it returns error, instead, i restarted the MGMT, using the command (svc_restart_service restart MGMT). The end result still displayed the alert.
For the third option, I could not clear the alert in the UI as there is no option to clear it. I also tried clearing using the commands you provided but also returns error as it is not listed in the svc_*.
Im not sure about the fourth option, I can only find that the firmware version for all DPE disks are LS5C.
I will try the basic "hardware advisory" as per recommended. However, could the error persist due to the system running system software version more than 180 days old? Also as for now, I would like to know whether it would disrupt business operation to confirm the severity of this alert.
DELL-Josh Cr
Community Manager
•
9792 Posts
•
45323 Points
0
0
Posted July 28th, 2026 11:44
Thank you for reaching out and for providing the detailed information and screenshot.
It sounds like you have successfully replaced both the SP A and SP B Battery Backup Units (BBUs), and the battery-level faults have cleared. However, the parent-level DPE (Disk Processor Enclosure) fault can sometimes become a stale or persistent alert that does not automatically clear even after the underlying subcomponent issues are resolved. This is a known behavior on VNXe systems.
Here are the recommended steps to try to clear this persistent DPE fault:
1. Verify No Other Subcomponent Faults Remain
2. Acknowledge and Clear Any Outstanding Alerts
3. Perform a Rolling Reboot of Both Storage Processors
A rolling reboot can often clear stale fault states at the DPE level:
4. Check the DPE Status After Reboot
After both SPs have been rebooted and are operating normally, check whether the DPE fault has cleared. In many cases, the rolling reboot will resolve the stale fault condition.