Log file Viewer
#mongodb logs for Friday the 8th of June, 2012
[00:17:37] <multi_io> ah, you have to add a some documents before enabling sharding, so the database can determine the range of the shard key values?
[01:59:40] <lqez> eka: Use bulk/batch insert function of driver. Like : http://php.net/manual/en/mongocollection.batchinsert.php
[02:00:29] <eka> lqez: I see... so no special thing... just pass a list of documents and that will do... in the case of python I meam
[02:02:30] <lqez> Yeap, I don't use PyMongo or other python mongo drivers now, but it'll be done by passing documents in list.
[02:19:26] <tester> Anyone know how to do a string != 'a' and string != 'b'??? i can only string: {$ne: 'a'} but not sure how to do 'b' also
[02:23:20] <lqez> tester: How about{ $and: [{string:{$ne:'a'}},{string:{$ne:'b'}}] } or { string: $ne: [ 'a', 'b' ] }
[04:00:44] <maodun> I just installed the mongodb ubuntu package - will it by default only accept connections from localhost? if not, how can I configure it to only talk to localhost?
[07:45:39] <henrykim> If it is true, it's quire weird because sometimes I recognized the percentages over 100.
[08:09:24] <ub|k> carsten: oh, right, sorry, i had read the advanced queries docs but the "2.0+" is just too small, i guess i missed it ;)
[13:51:13] <NodeX> I'll update the second post with some new quirks and tips I have picked up on my SOLR travels
[13:53:07] <NodeX> it doesn't scale all that well atm though .. you can cluster it but it's a bit of a pitfa
[14:06:29] <NodeX> The thing with solr when coming from a k/v or NoSQL store is you'll find the schema restrictive
[15:38:39] <solars> hey, quick question, I want to map/reduce and retrieve the last record (timestamp) in each group, therefore I have: https://gist.github.com/7eb028eec863bc1e56e5 - however, I think the timestamp comparison is somehow wrong, it works but I cannot find certain results, does anyone know whats wrong with this short snippet?
[15:48:33] <timkuijsten> is it possible to specify a per user configuration file for the mongo client? i'd like to store the passwords in it instead of specifying it on the command line
[15:50:40] <mw44118> Hi -- I have a question. I'm counting all the documents in a collection, and the query takes a really long time. I'm using mongoengine, which is a python wrapper for mongo, and I'm running this query: ms.MyCollection.objects.count()
[15:51:36] <mw44118> I understand that doing len(ms.MyCollection.objects()) is slow because it would pull ALL the data out of mongo, then make objects for them all, then count the number of elements in that list, but I don't understand why the .count() method is taking so long.
[16:00:08] <dgottlieb> mw44118: can you confirm that MongoEngine isn't doing that worst case scenario under the hood?
[16:08:13] <solars> can anyone tell me when map/reduce returns a value of null? I don't get that (the value attribute of the document in the returning collection)
[16:10:15] <mw44118> dgottlieb: I can not confirm that. If I wanted to get a count of documents in a collection at a mongo query, how would I do that?
[16:53:45] <mw44118> OK, it seems like mongoengine DOES do the worst possible thing when I run ms.MyCollection.objects.count(). Dangit.
[17:19:17] <bikcmp> I know I Had problems with something of mongodb wanting either 1 server or 3 servers.
[17:45:51] <elarson> does the $set update modifier only allow updating one field at a time? ie does {$set: {foo: "Foo", bar: "Bar"}} set the foo and bar attributes on the document?
[17:47:13] <elarson> sorry if I missed it in the docs. I didn't see it specified and I have an issue in a test where it looks like it doesn't update each field, just the last one (bar in the example above)
[17:58:53] <rhqq> hey, when i have dots in collection name, does mongo assumes each part as different namespace? like database_name.some.collection.name.4 will be 5 namespaces?
[18:08:07] <multi_io> can I query by value-of-array-element, e.g. db.mycoll.find({"arr[0].textattr": "foobar" }) ?
[18:10:06] <multi_io> (i.e. querying for documents with an attribute arr which is an array whose 0th element is a document with a string attribute named textattr and value "foobar")
[19:59:37] <avdeveloper> I'm pondering on denormalizing my data, but that would result in having redundant data across the collections
[20:00:21] <avdeveloper> thanks! I think I'll take a look at multiple collections vs embedded documents
[20:06:10] <avdeveloper> okay so I just finished reading embedding vs references. I understand how they work. Now I just wonder if it's a good practice to embed a select few attributes of the Player document such as name and id into a Game document so that fetching a Game data would be faster than having two separate calls?
[20:20:33] <bluesnow> Hi, I'm wondering how I would automatically start a MongoDB instance on a server when a Python web application starts.
[20:30:48] <bluesnow> It's going to be on a remote server, and the user just accesses it through a browser.
[20:31:03] <bluesnow> It's for internal use at a company, and I don't think the remote server will be on 24/7
[20:31:45] <bluesnow> Also, even if the remote server is on 24/7, the web app will only be used once a week maybe.
[20:32:33] <sirpengi> well, if the server is powered on anyway it doesn't make much different for the services to be on (but idle)
[20:33:52] <ranman> avdeveloper: yeah that's a pretty good use case, you don't have to put the whole player document in there
[20:36:19] <bluesnow> sirpengi: oh hmm, okay. I'll just see if I can keep the db instance running then
[20:51:32] <David[1]> Is just calling close() on the Connection instance the right way to close a database connection?
[21:23:26] <lolfrenz> is it possible to apply mapreduce only on documents matching certain criteria?
[21:44:49] <tystr> I'm sure this has been asked before, but in a collection of docuoments, what's the best way to ensure that embedded documents in the collection all have unique ids?
[22:10:17] <foobar2k> but almost all of the read operations get sent to the secondary node for some reason
[22:10:39] <foobar2k> is there some way to get read operations to balance better between our primary and secondary
[23:00:35] <m4nti5> imagine that I have k2 or and want to find k2 and k3, what query would you suggest me to do?