Skip to content
This repository has been archived by the owner on Nov 17, 2023. It is now read-only.

Flaky test on MacOS: test_dataset_take #17985

Closed
eric-haibin-lin opened this issue Apr 7, 2020 · 1 comment
Closed

Flaky test on MacOS: test_dataset_take #17985

eric-haibin-lin opened this issue Apr 7, 2020 · 1 comment

Comments

@eric-haibin-lin
Copy link
Member

test_gluon_data.test_dataset_take ... ok (0.0002s)
Bug: Gluon DataLoader terminates the process pool early while ... libc++abi.dylib: terminating with uncaught exception of type dmlc::Error: [02:24:07] ../src/storage/./cpu_shared_storage_manager.h:218: Check failed: count >= 0 (-2 vs. 0) : 


/Users/runner/runners/2.168.0/work/_temp/5da477cf-b82d-4965-ae8b-94a3551a8258.sh: line 1: 95004 Abort trap: 6           python3 -m nose --with-timer --verbose tests/python/unittest/ --exclude-test=test_extensions.test_subgraph --exclude-test=test_extensions.test_custom_op --exclude-test=test_gluon_data.test_recordimage_dataset_with_data_loader_multiworker

In PR #17466
https://github.com/apache/incubator-mxnet/pull/17466/checks?check_run_id=559681127

@leezu
Copy link
Contributor

leezu commented Apr 7, 2020

I suspect this is caused by the multiprocessing

2020-04-04T02:24:07.0618480Z test_gluon_data.test_multi_worker ... ok (0.4079s)
2020-04-04T02:24:07.1849230Z test_gluon_data.test_multi_worker_shape ... ok (0.1218s)
2020-04-04T02:24:07.7095350Z test_gluon_data.test_multi_worker_forked_data_loader ... ok (0.5233s)
2020-04-04T02:24:07.7120510Z test_gluon_data.test_multi_worker_dataloader_release_pool ... ok (0.0013s)

They are disabled now as of ff234db

Let's reopen this issue if the error occurs again.

@leezu leezu closed this as completed Apr 7, 2020
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

No branches or pull requests

2 participants