We have recently procured High-Performance Subnet Manager 1.2
The product is monitoring Infiniband Fabric
Does this device collect error logs from infiniband switches and if yes, how can we obtain a summary of these Infiniband errors.
I have the User Guide for this product but nothing is mentioned about this feature.
The guide talks about performance manager and a location /var/log!!
can somebody give me some idea on obtaining infiniband switches related logs from HSM.
This Subnet Manager does not collect error logs from the infiniband switches. The logs would indicate SM logs such as devices logging in and logging out of the infiniband fabric. The SM is used to service the Infiniband fabric like routing traffic, allowing devices to log in and out of the IB fabric and is not meant to be used to manage the switches.
There is a better tool if you want to look and troubleshoot the Infiniband fabric called FACT:
You can use this tool to collect switch logs and to a certain extent manage the switches.
Please let me know if you need any further help.