Has anyone experienced issue with deleting entities in datastore admin?

[A. Kong]

Hi all,

I used to be able to delete entities using the, obviously, 'Delete Entities' button in datastore admin.

However in last two days it simply stopped working. The kickoff jobs simply failed and stuck

The 'reason to retry' of these tasks are simply 'App Error'. 

 

How can I get the 'delete entities' to work again?

Here is some additional information:

We have defined a 'default' task queue which points to our 'app' module as target. This caused datastore operations failures.  Therefore I added a new a queue called 'data-backup' and set its target to 'ah-builtin-python-bundle' and used this taskqueue for datastore operations (delete, restore etc). In the last two days both queues ('default' and 'data-backup') are filled with failed kickoff jobs.

Cheers, 

Tony

[Patrice (Cloud Platform Support)]

Hi Tony,

It happens that certain mapreduce tasks just retry forever in circumstances like this. It's possible one job started and could never succeed, and it now blocks every other job you kickoff.

I would suggest manually stopping and emptying all your queues, then once they're down, you can go in your datastore viewer and manually edit all the "AE_" entries to have a Status of "aborted". This should prevent the back-end from trying them again.

Cheers