Closed sebrowne closed 2 months ago
The Autotester will be down for scheduled transition work starting 5:00 PM MDT 06/24/2024. Expecting resuming normal operations by 12:00 PM MDT 06/25/2024.
Scheduling merge of all ready-to-merge configurations: GCC, Clang, CUDA, CUDA-UVM
EDIT: Maintenance completed, builds appear to be working correctly.
The frequently-failing Adelus_vector_random tests are being re-disabled in the new CUDA configuration (#13158). That detail was lost during the transition (the 3-rank one was previously disabled). This should stop the CUDA build from halting PRs.
I will update the configuration wiki at https://github.com/trilinos/Trilinos/wiki/Pull-Request-Testing-Interface once the Intel build is transitioned. We will still need to update the Master Merge process, so there is a good chance that will fail this week.
EDIT: PR configurations have been updated on the wiki. Transition is complete other than the Python3 Framework unit tests.
There is a critical error with the backing TPL modules for the Intel configuration. A ticket is in to get the issue resolved, but it is likely that the autotester will not allow merges until the module is restored. Trilinos DevOps will mark that build as non-blocking to prevent major interruptions to development if it is not resolved by 9:00PM MDT 07/03/2024.
EDIT: The Intel PR/MM build has been set to non-blocking. SEMS is looking into the issue and has made some progress resolving it, but I doubt it will be completed before the end of the holiday weekend.
All transitions are completed, closing.
The Autotester will be undergoing significant configuration changes surrounding the end of June (06/30/2024). We will be modifying many configurations with the transition to RHEL8 from now until the beginning of July.
Please report any issues or ask any questions on this GitHub issue
Per-build status