Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Mellanox plugin should handle Bluefield card in DPU mode correctly #662

Open
ykulazhenkov opened this issue Mar 19, 2024 · 1 comment · May be fixed by #816
Open

Mellanox plugin should handle Bluefield card in DPU mode correctly #662

ykulazhenkov opened this issue Mar 19, 2024 · 1 comment · May be fixed by #816

Comments

@ykulazhenkov
Copy link
Collaborator

ykulazhenkov commented Mar 19, 2024

Mellanox plugin should handle Bluefield card in DPU mode correctly

Currently, the Mellanox plugin will try to change the FW configuration for the card and reboot the node.
Changing the FW configuration for BF in DPU mode requires a power cycle of the entire host.
The operator can't handle this correctly, which leads to an infinite reboot loop.

Proposed behavior:

  • The Mellanox plugin should fail if it detects that the FW configuration of the BF in DPU mode doesn't match the policy configuration (the operator will not proceed with the configuration)
  • The Mellanox plugin should skip the FW configuration reset for BF in DPU mode.

Alternative 1:

  • The Mellanox plugin should apply the configuration to the DPU's FW, print a SCARY WARNING to the log, and emit an event that a manual power cycle is required to break a reboot loop.

Alternative 2:

  • The Mellanox plugin should print a warning message if it detects that the DPU's FW config doesn't match the policy and then ignore the device.
  • The generic plugin should be modified to ignore BF devices in DPU mode if they have an invalid FW config.

I personally prefer the option from the "Proposed behavior" section.

@e0ne @SchSeba @zeeke Please, share your thoughts

@adrianchiris
Copy link
Collaborator

adrianchiris commented May 2, 2024

+1 for the Proposed behavior

hopefully for DPU we will be able to load configuration by mstfwreset+reboot and save the powercycle
then we can remove this check from the mellanox plugin

@e0ne e0ne linked a pull request Dec 13, 2024 that will close this issue
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

2 participants