ExaLogic: Data Collection For Troubleshooting Symbol Errors, Cabling, Switch Port Issues On Infiniband Switches
(Doc ID 2413588.1)
Last updated on APRIL 15, 2024
Applies to:
Oracle Exalogic Elastic Cloud Software - Version 2.0.6.0.0 and laterLinux x86-64
Oracle Solaris on x86-64 (64-bit)
Oracle Virtual Server x86-64
Goal
In Exalogic racks, we often see symbol errors, cabling or specific Switch port issues on Infiniband switches. These issues can lead to problems of Switch port(s) being automatically disabled.
This MOS Note provides detailed steps on data collection that is needed to troubleshoot and debug the Symbol errors, cabling and specific Switch port issues on Infiniband Switches.
Symbol errors are often caused due to bad or loose cabling, or specific connector port issues on an IB Switch or IB HCA card on the client Compute Node/Storage head.
To fix symbol errors one (or) more of following actions may be needed.
- Reseating the cables
- Replacing the cables
- Reconnecting the cables to other available connector ports on the IB Switches
- Reseating IB HCA card on the Client Compute Nodes or Storage heads.
- Replacing the IB HCA card on the Client Compute Nodes or Storage heads.
Exact corrective action to fix the issues can be determined by Support after reviewing the data collected by using steps in this MOS Note.
Solution
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
Goal |
Solution |
References |