After Restarting A Compute Node The EoIB And IPoIB Networks On vServers Started On That Node Fails To Initialize With Errors On The vServer's Console
(Doc ID 1519141.1)
Last updated on FEBRUARY 09, 2018
Applies to:Oracle Exalogic Elastic Cloud Software - Version 220.127.116.11.0 and later
Oracle Virtual Server (x86-64)
You have an Exalogic 18.104.22.168.0 Virtual deployment that is running 22.214.171.124.0 of the Exalogic Virtual Control Stack. At some point, possibly while troubleshooting an unrelated problem, you elect to:
- Fully stop the Exalogic Control Stack (in turn stopping the PC2, PC1, EC1, OVMM and DB components)
- Stop the Oracle Virtual Server (OVS) virtual machines that are running the above components of the Exalogic Control Stack
- Stop and reboot each of the compute nodes where the vServers running the Exalogic Control Stack reside
After the compute nodes complete restarting, you in turn restart the "vServers" running the Exalogic Control Stack in the order that is required but during this process you notice that, for a given VM, its EoIB and or IPoIB network addresses is no longer readily accessible, even though it ran successfully on this compute prior to the restart. When attempting a further re-start of one of the virtual machines, by adding a "-c" (to start a console during vm startup) you can see during system startup lines similar to the following on the affected "vServer's" console:
The problem may manifest when a "vServer" (Exalogic Control Stack or otherwise) is next started on a compute node that has recently been rebooted.
To view full details, sign in with your My Oracle Support account.
Don't have a My Oracle Support account? Click to get started!