-
Notifications
You must be signed in to change notification settings - Fork 380
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
support for Anvil in provenance.py was dropped #1417
Comments
I have branch that I am putting together for synchronizing mach_syslog logic across all of the systems. I'll put the Anvil support back in in this and then submit a PR. I would like to understand how Anvil support got dropped though. I also want this new PR to go into ACME directly, and not wait for it to go through CIME. Most of the changes will be in (what was) cime_config/acme/machines, but provenance.py will need to be changed as well. |
@worleyph probably got messed up in the ESMCI/ACME integration. I think the likelihood of these kinds of mistakes will decrease now that we are doing more frequent integrations. Sorry for the screwup. |
Other than the increased frequency, will the process be any different? Does not seem to be a frequency related issue. I assume that I need to wait for the CIME 5.3 integration to make it to master, and to start over, since everything has moved around? Thanks. |
@worleyph , increased frequency should lead to greatly reduced conflicts when we do merges and resolving conflicts is where mistakes can be made. Yes, it's probably wise to hold off on CIME changes until 5.3 merge is complete. It's already on next and should be on master soon. |
#1429) Update provenance data collection: a) Update mach_syslog for other platforms to match that for Anvil When adding mach_syslog for Anvil, the logic was refactored, decreasing the amount of data collected. Introduce similar changes for mach_syslog for the other systems. The system-specific commands are querying system and job status were also examined, and some of the command choices and options modified. b) Modify collection of provenance data on all supported systems Disable collection of xtdb2proc and mdiag system configuration data and qstat -f system load data or Titan. (May want to re-enable in the future, so just commenting out, not deleting.) Eliminate collection of some redundant or irrelevant date on SLURM systems at NERSC, and add collection of some data on system configuration and node usage by other jobs. Also fix error caused by changed name of ACME job script output on NERSC systems. Anvil support in provenance.py was lost inadvertently. Add it back. Also eliminate some unnecessary detail in the provenance data collected on Anvil. Eliminate some unnecessary detail in the provenance data collected on Mira and Cetus. [BFB] Fixes #1417
Update provenance data collection: a) Update mach_syslog for other platforms to match that for Anvil When adding mach_syslog for Anvil, the logic was refactored, decreasing the amount of data collected. Introduce similar changes for mach_syslog for the other systems. The system-specific commands are querying system and job status were also examined, and some of the command choices and options modified. b) Modify collection of provenance data on all supported systems Disable collection of xtdb2proc and mdiag system configuration data and qstat -f system load data or Titan. (May want to re-enable in the future, so just commenting out, not deleting.) Eliminate collection of some redundant or irrelevant date on SLURM systems at NERSC, and add collection of some data on system configuration and node usage by other jobs. Also fix error caused by changed name of ACME job script output on NERSC systems. Anvil support in provenance.py was lost inadvertently. Add it back. Also eliminate some unnecessary detail in the provenance data collected on Anvil. Eliminate some unnecessary detail in the provenance data collected on Mira and Cetus. [BFB] Fixes #1417
Update provenance data collection: a) Update mach_syslog for other platforms to match that for Anvil When adding mach_syslog for Anvil, the logic was refactored, decreasing the amount of data collected. Introduce similar changes for mach_syslog for the other systems. The system-specific commands are querying system and job status were also examined, and some of the command choices and options modified. b) Modify collection of provenance data on all supported systems Disable collection of xtdb2proc and mdiag system configuration data and qstat -f system load data or Titan. (May want to re-enable in the future, so just commenting out, not deleting.) Eliminate collection of some redundant or irrelevant date on SLURM systems at NERSC, and add collection of some data on system configuration and node usage by other jobs. Also fix error caused by changed name of ACME job script output on NERSC systems. Anvil support in provenance.py was lost inadvertently. Add it back. Also eliminate some unnecessary detail in the provenance data collected on Anvil. Eliminate some unnecessary detail in the provenance data collected on Mira and Cetus. [BFB] Fixes #1417
Add ability for create_test to infer non-default machine from tests Test suite: scripts_regression_tests --fast Test baseline: Test namelist changes: Test status: bit for bit Fixes #1417 User interface changes?: None Code review: @jedwards4b
Update provenance data collection: a) Update mach_syslog for other platforms to match that for Anvil When adding mach_syslog for Anvil, the logic was refactored, decreasing the amount of data collected. Introduce similar changes for mach_syslog for the other systems. The system-specific commands are querying system and job status were also examined, and some of the command choices and options modified. b) Modify collection of provenance data on all supported systems Disable collection of xtdb2proc and mdiag system configuration data and qstat -f system load data or Titan. (May want to re-enable in the future, so just commenting out, not deleting.) Eliminate collection of some redundant or irrelevant date on SLURM systems at NERSC, and add collection of some data on system configuration and node usage by other jobs. Also fix error caused by changed name of ACME job script output on NERSC systems. Anvil support in provenance.py was lost inadvertently. Add it back. Also eliminate some unnecessary detail in the provenance data collected on Anvil. Eliminate some unnecessary detail in the provenance data collected on Mira and Cetus. [BFB] Fixes #1417
Update provenance data collection: a) Update mach_syslog for other platforms to match that for Anvil When adding mach_syslog for Anvil, the logic was refactored, decreasing the amount of data collected. Introduce similar changes for mach_syslog for the other systems. The system-specific commands are querying system and job status were also examined, and some of the command choices and options modified. b) Modify collection of provenance data on all supported systems Disable collection of xtdb2proc and mdiag system configuration data and qstat -f system load data or Titan. (May want to re-enable in the future, so just commenting out, not deleting.) Eliminate collection of some redundant or irrelevant date on SLURM systems at NERSC, and add collection of some data on system configuration and node usage by other jobs. Also fix error caused by changed name of ACME job script output on NERSC systems. Anvil support in provenance.py was lost inadvertently. Add it back. Also eliminate some unnecessary detail in the provenance data collected on Anvil. Eliminate some unnecessary detail in the provenance data collected on Mira and Cetus. [BFB] Fixes #1417
Update needed for ACME Framework features brought in: * f444d0f Merge PR E3SM-Project#1418 'matthewhoffman/framework/output_record_reference_time' into develop * 263e14f Merge PR E3SM-Project#1428 'mark-petersen/framework/couple_fixes' into develop * bcce31d Merge PR E3SM-Project#1424 'amametjanov:az/tools/cp-prebuilt-tools' into develop * 98cfeea Merge PR# 1349 'akturner/framework/forcing_cleanup' into develop * 9359319 Merge PR E3SM-Project#1347 'akturner/framework/forcing_restart_timestamp' into develop * e9ce203 Merge PR E3SM-Project#1348 'akturner/framework/forcing_at_init' into develop * 4974284 Merge PR E3SM-Project#1368 'akturner/framework/improved_messages_in_driver' into develop * 86d50c5 Merge PR E3SM-Project#1417 'akturner/framework/forcing_multiple_blocks' into develop * 9116da3 Merge branch 'framework/validation-of-streams-using-interval_in-interval_out' into develop * e466b46 Merge branch 'framework/interval_in-interval_out-support-for-streams' into develop * 30dc955 Merge branch 'az/framework/mpas_dmpar-race-fix' into develop * b632938 Merge branch 'framework/i8_interval_division' into develop * 6dac06c Merge branch 'framework/log_write_IBM_error' into develop * 960a648 Merge branch 'framework/cleanup-logging-stream-manager' into develop * 504c282 Merge branch 'framework/make-streams-with-direction-none-inactive' into develop * 5903748 Merge branch 'framework/correctly_remove_blk_fields' into develop * 3565965 Merge branch 'framework/iostreams-real4dfield-bug' into develop * 8b60591 Merge branch 'framework/missing-deallocate-nEdgesOnCellField-bootstrapping' into develop * 70b953b Merge branch 'master' into develop
PR #1248 added support for calling mach_syslog and collecting other provenance-related data for Anvil to provenance.py. This has since disappeared, and it is not in CIME 5.3 either. I would appreciate that this be restored. I can submit another PR if required but would like this to go in quickly.
The text was updated successfully, but these errors were encountered: