Detecting Run Failures and Machine Crashes in MinKNOW

[irisxu-CPOS]

We are building an automated nanopore data processing pipeline and would like advice on detecting failures programmatically.

Specifically:

• Are there recommended ways in MinKNOW to real-time detect run failures or error states for automated monitoring?
• How can machine-level failures be detected? We have encountered cases where the machine powered off or crashed during run, causing downstream data transfer failed even though data and reports were generated locally.

Any guidance or documentation would be greatly appreciated.

[0x55555555]

Hello,

Have you reviewed the examples included in the repository?

https://github.com/nanoporetech/minknow_api/blob/master/python/minknow_api/examples/export_to_csv.py looks like it may help with what you're trying to achieve.

MinKNOW API itself does not contain any information about the machine failing itself - you would need to run direct OS queries.

Thanks,

• George