Dear HPC Users,
We are experiencing a critical issue with the login node on our HPC cluster, where the server is crashing intermittently. Our IPMI logs indicate an “OS Critical Stop” error, and we suspect this may be caused by a software or application running on the login node that is either outdated or self-compiled.
Due to the nature of these immediat crashes, there are no system logs that can help us identify the responsible.
To help us resolve this issue as soon as possible, we request that:
- If you believe you may have run such software (even as a test), please contact us immediately.
- If you suspect your software may have triggered this issue, please stop running it and notify us right away with details.
Your prompt response is crucial to ensuring the stability and availability of the HPC system for all users.
We appreciate your cooperation in this matter.
Best regards,