ALERT: Constant Node Panics or Tight Boot Cycle Looping Immediately After an ODA Bare Metal 18.7 (DCS) Upgrade or CPU Reduction after Deployment. DCS-10001 "Configured core count is not equal maximum allowed core count"
(Doc ID 2599834.1)
Last updated on JULY 20, 2024
Applies to:
Oracle Database Appliance Software - Version 18.3.0.0.0 ODAVP to 18.7.0.0.0 Bare Metal [Release 12.2]Information in this document applies to any platform.
Symptoms
After an ODA Upgrade 18.7 on Bare Metal (DCS stack) this problem can have one or more of the following symptoms:
- Tight OS boot cycling preventing the node from starting and staying up successfully.
- The patched node will restart and get caught in a tight node reboot cycle which can occur within one minute after the prior OS startup process.
Other symptoms can include
- DCS-10001:Internal error encountered: Configured core count is not equal maximum allowed core count.
- Constant Node Panics Immediately after reboot.
NOTE: This problem does not exist before or after 18.7 on the ODA.
18.8 is available and is the recommended upgrade path if using any 18.x ODA bare metal version.
The problem described in this ALERT for 18.7 is avoided by upgrading to 18.8.
Most importantly, 18.8 is also the required 18.x patchset level if upgrading to 19.x.
18.8 is available and is the recommended upgrade path if using any 18.x ODA bare metal version.
The problem described in this ALERT for 18.7 is avoided by upgrading to 18.8.
Most importantly, 18.8 is also the required 18.x patchset level if upgrading to 19.x.
In order to encounter this problem you must have the following configuration:
- Use the DCS Stack (Bare Metal X4-2 up to X8-2*)
- Use less than the total number of hardware cpus available on the ODA hardware
- Be on 18.7 (
* The CPU count available for the ODA will vary depending on the hardware used.
Changes
Encountering this problem requires:
- An upgrade or deploy the ODA to DCS stack 18.7
and - There must be an ODA CPU Core reduction either before or after the 18.7 bundle is being used and
- The fix for Bug 30313635 is not applied.
An Oracle Database Appliance system is provisioned with all CPU cores enabled by default.
The core reduction can be due to either of the following:
- ODA configurations which have previously altered the CPU Core Count to a lower value than the hardware type maximum before patching.
- ODA configuration which alter the CPU core count to a lower value than the hardware type maximum after patching.
Cause
To view full details, sign in with your My Oracle Support account. |
|
Don't have a My Oracle Support account? Click to get started! |
In this Document
| Symptoms |
| Changes |
| Cause |
| Solution |
| References |