The question is as basic as it is simple... How do you log all queries in a "tail"able log file in mongodb?

I have tried:

  • setting the profiling level
  • setting the slow ms parameter starting
  • mongod with the -vv option

The /var/log/mongodb/mongodb.log keeps showing just the current number of active connections...

Solution 1

You can log all queries:

$ mongo
MongoDB shell version: 2.4.9
connecting to: test
> use myDb
switched to db myDb
> db.getProfilingLevel()
0
> db.setProfilingLevel(2)
{ "was" : 0, "slowms" : 1, "ok" : 1 }
> db.getProfilingLevel()
2
> db.system.profile.find().pretty()

Source: http://docs.mongodb.org/manual/reference/method/db.setProfilingLevel/

db.setProfilingLevel(2) means "log all operations".

Solution 2

I ended up solving this by starting mongod like this (hammered and ugly, yeah... but works for development environment):

mongod --profile=1 --slowms=1 &

This enables profiling and sets the threshold for "slow queries" as 1ms, causing all queries to be logged as "slow queries" to the file:

/var/log/mongodb/mongodb.log

Now I get continuous log outputs using the command:

tail -f /var/log/mongodb/mongodb.log

An example log:

Mon Mar  4 15:02:55 [conn1] query dendro.quads query: { graph: "u:http://example.org/people" } ntoreturn:0 ntoskip:0 nscanned:6 keyUpdates:0 locks(micros) r:73163 nreturned:6 reslen:9884 88ms

Solution 3

Because its google first answer ...
For version 3

$ mongo
MongoDB shell version: 3.0.2
connecting to: test
> use myDb
switched to db
> db.setLogLevel(1)

http://docs.mongodb.org/manual/reference/method/db.setLogLevel/

Solution 4

MongoDB has a sophisticated feature of profiling. The logging happens in system.profile collection. The logs can be seen from:

db.system.profile.find()

There are 3 logging levels (source):

  • Level 0 - the profiler is off, does not collect any data. mongod always writes operations longer than the slowOpThresholdMs threshold to its log. This is the default profiler level.
  • Level 1 - collects profiling data for slow operations only. By default slow operations are those slower than 100 milliseconds. You can modify the threshold for slow operations with the slowOpThresholdMs runtime option or the setParameter command. See the Specify the Threshold for Slow Operations section for more information.
  • Level 2 - collects profiling data for all database operations.

To see what profiling level the database is running in, use

db.getProfilingLevel()

and to see the status

db.getProfilingStatus()

To change the profiling status, use the command

db.setProfilingLevel(level, milliseconds)

Where level refers to the profiling level and milliseconds is the ms of which duration the queries needs to be logged. To turn off the logging, use

db.setProfilingLevel(0)

The query to look in the system profile collection for all queries that took longer than one second, ordered by timestamp descending will be

db.system.profile.find( { millis : { $gt:1000 } } ).sort( { ts : -1 } )

Solution 5

I made a command line tool to activate the profiler activity and see the logs in a "tail"able way --> "mongotail":

$ mongotail MYDATABASE
2015-02-24 19:17:01.194 QUERY  [Company] : {"_id": ObjectId("548b164144ae122dc430376b")}. 1 returned.
2015-02-24 19:17:01.195 QUERY  [User] : {"_id": ObjectId("549048806b5d3db78cf6f654")}. 1 returned.
2015-02-24 19:17:01.196 UPDATE [Activation] : {"_id": "AB524"}, {"_id": "AB524", "code": "f2cbad0c"}. 1 updated.
2015-02-24 19:17:10.729 COUNT  [User] : {"active": {"$exists": true}, "firstName": {"$regex": "mac"}}
...

But the more interesting feature (also like tail) is to see the changes in "real time" with the -f option, and occasionally filter the result with grep to find a particular operation.

See documentation and installation instructions in: https://github.com/mrsarm/mongotail

Solution 6

if you want the queries to be logged to mongodb log file, you have to set both the log level and the profiling, like for example:

db.setLogLevel(1)
db.setProfilingLevel(2)

(see https://docs.mongodb.com/manual/reference/method/db.setLogLevel)

Setting only the profiling would not have the queries logged to file, so you can only get it from

db.system.profile.find().pretty()

Solution 7

Once profiling level is set using db.setProfilingLevel(2).

The below command will print the last executed query.
You may change the limit(5) as well to see less/more queries.
$nin - will filter out profile and indexes queries
Also, use the query projection {'query':1} for only viewing query field

db.system.profile.find(
{ 
    ns: { 
        $nin : ['meteor.system.profile','meteor.system.indexes']
    }
} 
).limit(5).sort( { ts : -1 } ).pretty()

Logs with only query projection

db.system.profile.find(
{ 
    ns: { 
        $nin : ['meteor.system.profile','meteor.system.indexes']
    }
},
{'query':1}
).limit(5).sort( { ts : -1 } ).pretty()

Solution 8

The profiler data is written to a collection in your DB, not to file. See http://docs.mongodb.org/manual/tutorial/manage-the-database-profiler/

I would recommend using 10gen's MMS service, and feed development profiler data there, where you can filter and sort it in the UI.

Solution 9

I think that while not elegant, the oplog could be partially used for this purpose: it logs all the writes - but not the reads...

You have to enable replicatoon, if I'm right. The information is from this answer from this question: How to listen for changes to a MongoDB collection?

Solution 10

Setting profilinglevel to 2 is another option to log all queries.

Solution 11

db.setProfilingLevel(2,-1)

This worked! it logged all query info in mongod log file

Solution 12

I recommend checking out mongosniff. This can tool can do everything you want and more. Especially it can help diagnose issues with larger scale mongo systems and how queries are being routed and where they are coming from since it works by listening to your network interface for all mongo related communications.

http://docs.mongodb.org/v2.2/reference/mongosniff/

Solution 13

I wrote a script that will print out the system.profile log in real time as queries come in. You need to enable logging first as stated in other answers. I needed this because I'm using Windows Subsystem for Linux, for which tail still doesn't work.

https://github.com/dtruel/mongo-live-logger

Solution 14

db.adminCommand( { getLog: "*" } )

Then

db.adminCommand( { getLog : "global" } )

Solution 15

This was asked a long time ago but this may still help someone:

MongoDB profiler logs all the queries in the capped collection system.profile. See this: database profiler

  1. Start mongod instance with --profile=2 option that enables logging all queries OR if mongod instances is already running, from mongoshell, run db.setProfilingLevel(2) after selecting database. (it can be verified by db.getProfilingLevel(), which should return 2)
  2. After this, I have created a script which utilises mongodb's tailable cursor to tail this system.profile collection and write the entries in a file. To view the logs I just need to tail it:tail -f ../logs/mongologs.txt. This script can be started in background and it will log all the operation on the db in the file.

My code for tailable cursor for the system.profile collection is in nodejs; it logs all the operations along with queries happening in every collection of MyDb:

const MongoClient = require('mongodb').MongoClient;
const assert = require('assert');
const fs = require('fs');
const file = '../logs/mongologs'
// Connection URL
const url = 'mongodb://localhost:27017';

// Database Name
const dbName = 'MyDb';
//Mongodb connection

MongoClient.connect(url, function (err, client) {
   assert.equal(null, err);
   const db = client.db(dbName);
   listen(db, {})
});

function listen(db, conditions) {
var filter = { ns: { $ne: 'MyDb.system.profile' } }; //filter for query
//e.g. if we need to log only insert queries, use {op:'insert'}
//e.g. if we need to log operation on only 'MyCollection' collection, use {ns: 'MyDb.MyCollection'}
//we can give a lot of filters, print and check the 'document' variable below

// set MongoDB cursor options
var cursorOptions = {
    tailable: true,
    awaitdata: true,
    numberOfRetries: -1
};

// create stream and listen
var stream = db.collection('system.profile').find(filter, cursorOptions).stream();

// call the callback
stream.on('data', function (document) {
    //this will run on every operation/query done on our database
    //print 'document' to check the keys based on which we can filter
    //delete data which we dont need in our log file

    delete document.execStats;
    delete document.keysExamined;
    //-----
    //-----

    //append the log generated in our log file which can be tailed from command line
    fs.appendFile(file, JSON.stringify(document) + '\n', function (err) {
        if (err) (console.log('err'))
    })

});

}

For tailable cursor in python using pymongo, refer the following code which filters for MyCollection and only insert operation:

import pymongo
import time
client = pymongo.MongoClient()
oplog = client.MyDb.system.profile
first = oplog.find().sort('$natural', pymongo.ASCENDING).limit(-1).next()

ts = first['ts']
while True:
    cursor = oplog.find({'ts': {'$gt': ts}, 'ns': 'MyDb.MyCollection', 'op': 'insert'},
                        cursor_type=pymongo.CursorType.TAILABLE_AWAIT)
    while cursor.alive:
        for doc in cursor:
            ts = doc['ts']
            print(doc)
            print('\n')
        time.sleep(1)

Note: Tailable cursor only works with capped collections. It cannot be used to log operations on a collection directly, instead use filter: 'ns': 'MyDb.MyCollection'

Note: I understand that the above nodejs and python code may not be of much help for some. I have just provided the codes for reference.

Use this link to find documentation for tailable cursor in your languarge/driver choice Mongodb Drivers

Another feature that i have added after this logrotate.

Solution 16

Try out this package to tail all the queries (without oplog operations): https://www.npmjs.com/package/mongo-tail-queries

(Disclaimer: I wrote this package exactly for this need)