Skip to content

SW1 on london stuck in thermal shutdown #2060

Open
@Aaron-Hartwig

Description

@Aaron-Hartwig
Contributor

As reported in various places in chat. Since the FPGA handles the Tofino sequencing and can survive SP reset, we've had bugs in the past where the SP and FPGA get out of sync with one another and do not recover properly. We expect that may have been the case here somehow but could not verify that since the production images don't have udprpc, rendering hiffy unusable in this context.

I've flashed the v1.0.37 dev image on this switch (which has udprpc) so we can debug further if it reproduces.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

      Development

      No branches or pull requests

        Participants

        @Aaron-Hartwig

        Issue actions

          `SW1` on `london` stuck in thermal shutdown · Issue #2060 · oxidecomputer/hubris