More debug

Author: carmocca

.github/workflows/ci-pytorch-test-tpu.yml

@@ -78,43 +78,40 @@ jobs:
 
     - name: Deploy cluster
       run: |
-       export PATH=$PATH:$HOME/go/bin
-       job_name=$(jsonnet -J ml-testing-accelerators/ dockers/tpu-tests/tpu_test_cases.jsonnet | kubectl create -f -)
-       job_name=${job_name#job.batch/}
-       job_name=${job_name% created}
-       pod_name=$(kubectl get po -l controller-uid=`kubectl get job $job_name -o "jsonpath={.metadata.labels.controller-uid}"` | awk 'match($0,!/NAME/) {print $1}')
-       echo "GKE pod name: $pod_name"
-       echo "Waiting on kubernetes job: $job_name"
-       status_code=2 &&
-       # Check on the job periodically. Set the status code depending on what happened to the job in Kubernetes.
-       printf "Waiting for job to finish: "
-       while true; do
-         if kubectl get jobs $job_name -o jsonpath='Failed:{.status.failed}' | grep "Failed:1"; then
-           status_code=1 && break;
-         elif kubectl get jobs $job_name -o jsonpath='Succeeded:{.status.succeeded}' | grep "Succeeded:1"; then
-           status_code=0 && break;
-         else
-           printf ".";
-         fi;
-         sleep 5;
-       done
-       echo "Done waiting. Job status code: $status_code"
-       kubectl logs -f $pod_name --container=train > /tmp/full_output.txt
-       cat /tmp/full_output.txt
-       if grep -q '<?xml version="1.0" ?>' /tmp/full_output.txt; then
-         csplit /tmp/full_output.txt '/<?xml version="1.0" ?>/';
-       else
-         mv /tmp/full_output.txt xx00;
-       fi
-       # First portion is the test logs.
-       cat xx00 && echo "Done with log retrieval attempt."
-       exit $status_code
+        export PATH=$PATH:$HOME/go/bin
+        job_name=$(jsonnet -J ml-testing-accelerators/ dockers/tpu-tests/tpu_test_cases.jsonnet | kubectl create -f -)
+        job_name=${job_name#job.batch/}
+        job_name=${job_name% created}
+        pod_name=$(kubectl get po -l controller-uid=`kubectl get job $job_name -o "jsonpath={.metadata.labels.controller-uid}"` | awk 'match($0,!/NAME/) {print $1}')
+        echo "GKE pod name: $pod_name"
+        echo "Waiting on kubernetes job: $job_name"
+        status_code=2 &&
+        # Check on the job periodically. Set the status code depending on what happened to the job in Kubernetes.
+        printf "Waiting for job to finish: "
+        while true; do
+          if kubectl get jobs $job_name -o jsonpath='Failed:{.status.failed}' | grep "Failed:1"; then
+            status_code=1 && break;
+          elif kubectl get jobs $job_name -o jsonpath='Succeeded:{.status.succeeded}' | grep "Succeeded:1"; then
+            status_code=0 && break;
+          else
+            printf ".";
+          fi;
+          sleep 5;
+        done
+        echo "Done waiting. Job status code: $status_code"
+        kubectl logs -f $pod_name --container=train > /tmp/full_output.txt
+        grep '<?xml version="1.0" ?>' /tmp/full_output.txt  # sanity check
+        csplit /tmp/full_output.txt '/<?xml version="1.0" ?>/'
+        # REVERT ME
+        ls
+        cat xx01
+        exit $status_code
       shell: bash
 
     - name: Statistics
       if: success()
       run: |
-        mv ./xx01 coverage.xml
+        mv xx01 coverage.xml
         pip install coverage -q
         coverage report
         coverage xml