Skip to content
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

dlrover/blob/master/docs/tutorial/tf_elasticjob_on_k8s 【tf_elasticjob_on_k8s example failed to start】 【tf_elasticjob_on_k8s 示例启动失败】 #1121

Closed
lichadehehehe opened this issue May 14, 2024 · 5 comments
Assignees

Comments

@lichadehehehe
Copy link

while executing the tf_elasticjob_on_k8s example, job fails with error.

cd examples/tensorflow/criteo_deeprec
kubectl apply -f autoscale_job.yaml

initial error messages as follows:

[root@localhost test]# kubectl get pods -n dlrover
NAME                                           READY   STATUS    RESTARTS   AGE
deepctr-auto-scale-edljob-chief-0              0/1     Error     0          15h
deepctr-auto-scale-edljob-evaluator-0          0/1     Error     0          15h
deepctr-auto-scale-edljob-ps-0                 0/1     Error     0          15h
dlrover-controller-manager-687c846bb-95bst     2/2     Running   0          23h
elasticjob-deepctr-auto-scale-dlrover-master   0/1     Error     0          15h 

error logs from the chief pod
kubectl logs -f deepctr-auto-scale-edljob-chief-0 -n dlrover

 2024-05-13 19:13:25.419	
/bin/bash: line 0: cd: /home/examples/tensorflow/criteo_deeprec: No such file or directory

upon unpacking the image of registry.cn-hangzhou.aliyuncs.com/intell-ai/dlrover:deeprec_criteo_v1 it appears that the directory is missing. however there are files in the directory of /dlrover/examples/tensorflow/criteo_deeprec

image

I've changed the spec.containers.command of the examples/tensorflow/criteo_deeprec/autoscale_job.yaml/ file as follows for trying to remedy the issue:

              command:
                - /bin/bash
                - -c
                - "cd /dlrover/examples/tensorflow/criteo_deeprec \
                  && python -m dlrover.trainer.entry.local_entry \
                  --platform=Kubernetes --conf=train_conf.TrainConf \
                  --enable_auto_scaling=True"

the failure persists with the following error logs:

dlrover-master error logs:
image

chief error logs:
image

@workingloong
Copy link
Collaborator

I have fixed the example in #1141 . You can try the example with the latest master branch.

@workingloong workingloong assigned workingloong and unassigned hxdtest May 28, 2024
@lichadehehehe
Copy link
Author

lichadehehehe commented May 28, 2024

[2024-05-28 13:44:00,597] [WARNING][estimator_util.py:172:_unique] GlobalStepHook has existed, it won't be added
WARNING:dlrover.trainer:GlobalStepHook has existed, it won't be added
[2024-05-28 13:44:00,597] [WARNING][estimator_util.py:172:_unique] ElasticDataShardReportHook has existed, it won't be added
WARNING:dlrover.trainer:ElasticDataShardReportHook has existed, it won't be added
[2024-05-28 13:44:00,598] [WARNING][estimator_util.py:172:_unique] ReportModelInfoHook has existed, it won't be added
WARNING:dlrover.trainer:ReportModelInfoHook has existed, it won't be added
[2024-05-28 13:44:00,598] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,598] [INFO][estimator_util.py:180:append_hooks] Appending hooks after deduplication: training_hooks = ['GlobalStepHook', 'ElasticDataShardReportHook', 'ReportModelInfoHook', 'StopAtStepHook']
INFO:dlrover.trainer:Appending hooks after deduplication: training_hooks = ['GlobalStepHook', 'ElasticDataShardReportHook', 'ReportModelInfoHook', 'StopAtStepHook']
[2024-05-28 13:44:00,598] [INFO][estimator_util.py:162:append_hooks] Hooks before deduplication: training_chief_hooks = ['CheckpointSaverHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook']
INFO:dlrover.trainer:Hooks before deduplication: training_chief_hooks = ['CheckpointSaverHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook', 'StopAtStepHook']
[2024-05-28 13:44:00,598] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,598] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,598] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,598] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,598] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,598] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,598] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,598] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,599] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,599] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,599] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,599] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,599] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,599] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,599] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,599] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,599] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,599] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,599] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added

something's not quite right, I'm not sure what would have caused it...

@lichadehehehe
Copy link
Author

Every 2.0s: kubectl get pods -n dlrover                                                                                                                                    10-202-91-11: Tue May 28 14:01:41 2024

NAME                                           READY   STATUS    RESTARTS   AGE
deepctr-auto-scale-edljob-chief-0              1/1     Running   0          130m
deepctr-auto-scale-edljob-evaluator-0          1/1     Running   0          130m
deepctr-auto-scale-edljob-ps-0                 1/1     Running   0          130m
dlrover-controller-manager-5d79b766f6-t9jcq    2/2     Running   0          159m
elasticjob-deepctr-auto-scale-dlrover-master   1/1     Running   0          130m

it appears that workers are failed to start

@workingloong
Copy link
Collaborator

[2024-05-28 13:44:00,598] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,598] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,598] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added
[2024-05-28 13:44:00,598] [WARNING][estimator_util.py:172:_unique] StopAtStepHook has existed, it won't be added
WARNING:dlrover.trainer:StopAtStepHook has existed, it won't be added

it does not matter.

@workingloong
Copy link
Collaborator

Every 2.0s: kubectl get pods -n dlrover                                                                                                                                    10-202-91-11: Tue May 28 14:01:41 2024

NAME                                           READY   STATUS    RESTARTS   AGE
deepctr-auto-scale-edljob-chief-0              1/1     Running   0          130m
deepctr-auto-scale-edljob-evaluator-0          1/1     Running   0          130m
deepctr-auto-scale-edljob-ps-0                 1/1     Running   0          130m
dlrover-controller-manager-5d79b766f6-t9jcq    2/2     Running   0          159m
elasticjob-deepctr-auto-scale-dlrover-master   1/1     Running   0          130m

it appears that workers are failed to start

You can use kubectl to delete a worker and a new work will restart.
kubectl -n dlrover delete pod deepctr-auto-scale-edljob-worker-0

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants