I'm looking for advice on how to efficiently query hardware health data from around 190 HPE iLOs. My current method using the HPE PowerShell module is too slow, taking about 20 seconds per server, which can add up to over an hour. I've tried working with the Redfish API and RIBCL, but I'm still facing delays. Unfortunately, I can't use SNMP at the moment since I'm in the middle of a larger project to build a collector. I'm in urgent need of this data, so I wonder if anyone can share their experiences or suggest ways to optimize the process. Would parallel execution be my best bet, or are there other methods I could consider to speed things up? Thanks for taking the time to read this!
5 Answers
Have you thought about running your queries in parallel? Instead of waiting for one server to finish before moving on to the next, you can poll multiple iLOs at the same time. This can drastically cut down the total processing time. Just be careful with network bandwidth if you're polling too many at once. There are a ton of tutorials out there on how to implement this with PowerShell using jobs or runspaces. Definitely worth checking out!
Definitely go with async or parallel execution. You could set it up as a scheduled task that exports a summary. This approach should save you a lot of time!
Thanks! Sounds like a solid strategy, and I appreciate the suggestion.
You might want to give HP OneView a shot if you haven’t already. It could simplify the process depending on your setup.
I ran into issues with HPE servers and ESXi before. I remember having to manually set both the iLO and the switch to 100/Full to avoid performance issues. If you're stuck at 20 seconds per server with Redfish, it might be worth checking the network settings and iLO configurations. I used PowerShell to query various HP systems before, and I was getting responses in seconds, so it’s definitely possible to speed it up if the setup is right!
Using the Redfish API in parallel would be the way to go for your needs. There are tons of resources on PowerShell parallel execution that can help. Start testing with a few iLOs and see how it performs before scaling.
Thanks for your input! I'll definitely look into the parallel processing options. I just want to make sure it doesn’t overload the network.