Is it possible to create an aggregation in Pymongo

Question

The aggregation looks like:

res = Things.objects._collection.aggregate(pipeline,
            allowDiskUse=True,
            batchSize=101,
            noCursorTimeout=True
        )

and fails with

OperationFailure: unrecognized field 'noCursorTimeout'

It appears from the Pymongo documentation that no_cursor_timeout can be used with find(), however I have been unsuccessful locating a comparable **kwargs for aggregate().

I do not want to use setParameter or maxTimeMS.

UPDATE:

From MongoDB's Jira Site: SERVER-15042

The NoCursorTimeout bit in the OP_QUERY header prevents the server from closing a cursor that's idle for more than 10 minutes. Clients can set this bit if they may spend more than 10 minutes processing a batch of results.

The "aggregate" and "parallelCollectionScan" commands should allow the client to turn off cursor timeouts, too.

The response to this ticket was:

We intentionally do not support noCursorTimeout. The rationale for needing this will go away when we implement cursor keepalive. That work will be tracked in SERVER-6036.

Looking at SERVER-6036:

• Updated: Sep 14 2016 03:53:05 PM GMT+0000
• Status: OPEN
• Resolution: Unresolved
• Fix Version/s: planned but not scheduled

If I have read the documents correctly, a cursor no timeout option does not exist yet for aggregate().

Answer 1

You're correct, cursor timeout can't be disabled on individual aggregation cursors. But we've provided a global config setting, cursorTimeoutMillis, that you can increase to a large number:

mongod --setParameter cursorTimeoutMillis=600000  # 10 minutes

See SERVER-8188 for MongoDB version info, and how to set this parameter at runtime.