matensemble 0.4.2__tar.gz → 0.4.3__tar.gz

{matensemble-0.4.2 → matensemble-0.4.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: matensemble
-Version: 0.4.2
+Version: 0.4.3
 Summary: An adaptive and highly asynchronous ensemble simulation workflow manager MatEnsemble (https://github.com/Q-CAD/MatEnsemble) built jointly on top of the hierarchical graph based scheduler FLUX and concurrent-futures infrastructure of python
 Author: Soumendu Bagchi, Kaleb Duchesneau
 Author-email: Soumendu Bagchi <soumendubagchi@gmail.com>, Kaleb Duchesneau <kalebduchesneau@gmail.com>

{matensemble-0.4.2 → matensemble-0.4.3}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "matensemble"
-version = "0.4.2"
+version = "0.4.3"
 description = "An adaptive and highly asynchronous ensemble simulation workflow manager MatEnsemble (https://github.com/Q-CAD/MatEnsemble) built jointly on top of the hierarchical graph based scheduler FLUX and concurrent-futures infrastructure of python"
 readme = "README.md"
 license-files = ["LICENSE"]

{matensemble-0.4.2 → matensemble-0.4.3}/src/matensemble/TODO.md

@@ -131,12 +131,12 @@
 - [x] If that doesn't work break it down into smaller pieces
   ## --- Smaller pieces ---
   - [x] Maybe start with an nvidia image rather than Neil's image
-  - [ ] Make sure that flux works in the container
-  - [ ] Create a container that just has flux and have some different tests for that
-  - [ ] Create a container that has just MPI and test that make sure it works
-  - [ ] Combine flux and MPI and see if that works
-  - [ ] Create a container that has lammps and make sure that that is working
-  - [ ] Combine all the pieces
+  - [x] Make sure that flux works in the container
+  - [x] Create a container that just has flux and have some different tests for that
+  - [x] Create a container that has just MPI and test that make sure it works
+  - [x] Combine flux and MPI and see if that works
+  - [x] Create a container that has lammps and make sure that that is working
+  - [x] Combine all the pieces
 
 ## --- Test Frontier Apptainer container ---
 - [x] Need lots more help here
@@ -227,7 +227,7 @@
 - [x] Patch MatEnsemble to ignore the double free error: 134 or 137 idk whichever one it is
 - [x] Update the Base image or retag the one that I used to build this john
 - [x] The double free is likely from symmetrix so forget fixing it
-- [ ] Run the release script and do a PR
+- [x] Run the release script and do a PR
 
 ## --- Demo ---
 - [ ] Video demonstrating the MCP server

{matensemble-0.4.2 → matensemble-0.4.3}/src/matensemble/strategy.py

@@ -191,7 +191,11 @@ class NonAdaptiveStrategy(FutureProcessingStrategy):
                 had_failure = True
                 continue
 
-            if rc != 0:
+            # rc 134 is a double free or corruption error caused by lammps-symmetrix in the
+            # in the frontier image during lammps cleanup
+            # the function will still complete successfully and produce a result.pickle file
+            # so if we can safely ignore the return code
+            if rc != 0 and rc != 134:
                 append_text(
                     chore.workdir / "stderr",
                     f"\n\n===== MATENSEMBLE: NONZERO EXIT =====\nchore={chore_id} rc={rc}\n",
@@ -283,7 +287,11 @@ class UserStrategy(FutureProcessingStrategy):
                 had_failure = True
                 continue
 
-            if rc != 0:
+            # rc 134 is a double free or corruption error caused by lammps-symmetrix in the
+            # in the frontier image during lammps cleanup
+            # the function will still complete successfully and produce a result.pickle file
+            # so if we can safely ignore the return code
+            if rc != 0 and rc != 134:
                 append_text(
                     chore.workdir / "stderr",
                     f"\n\n===== MATENSEMBLE: NONZERO EXIT =====\nchore={chore_id} rc={rc}\n",