It is a new syndication framework framework for data aggregation, which is similar in concept to the pipeline data processing. Each document through a pipeline by a plurality of nodes, each node has its own special features (grouping, filtering, etc.), the document after the pipeline processing, the final output of the results.
The pipeline has two basic functions:
First, the document "filter", which is screened qualifying documents;
Second, the document "transformation" that is, change the output in the form of documents.
errmsg: Exceeded memory limit for $ group, but did not allow external Sort.Pass allowDiskuse: true
I use this data table is: there are about four million data
I have to check that: You receive information up to the first five users
Written statement analysis, in strict accordance with written grammar;
Since there is a problem, we have to look Debug (we have code agricultural force)
If the pipe does not give the expected results, you need to debug, debugging, you can specify only the first pipeline operator, and if so get the desired results, then it specifies the second pipeline operator, said the example above , first try $ project operator polymerization, if this operator results are valid, just add $ group operator,
If the result is still valid, add $ sort, and finally add $ limit operator, so that you can gradually cause the problem to locate the operator.
Stepping found no problems are not grammatical errors
Ma eggs, because the results of the polymerization must be limited to 16M operation (the maximum supported size mongodb impact of information), or you must do in the disk cache (allowDiskUse = True).
Finally, the statistical data you want to engage in out, like Ben & Poor's big.