Log file Viewer
#mongodb logs for Saturday the 26th of July, 2014
[00:31:35] <a|3xxx> i use the copydb command, once it completes, my user doesn't have access to that new database
[02:08:23] <LouisT> remember, you have to put mongodb in no auth, then connect with "mongo", then run: db.getSiblingDB("admin").runCommand({authSchemaUpgrade: 1 });
[02:37:35] <LouisT> from what i can tell it has a 16MB line limit, but i'm not completely sure about your question
[02:43:18] <LouisT> not exactly sure, is there a reason you're using mongoexport/mongoimport instead of mongodump/mongorestore?
[02:47:22] <s2013> https://gist.github.com/ss2k/611f45af83fdcfe4baa9 this is how the format looks like
[02:48:01] <s2013> i removed the results part so it was just a bunch of records and it imported one collection but says its too large for others
[02:49:03] <s2013> so how would yourecommend ? i can write a script that does each one individually but im sure thats the longest way to do it
[02:50:16] <s2013> well no.. parswe has shitty support. in fact it has nos upport.. we pay $2k/mo to them for no support
[02:56:34] <a|3xxx> LouisT, no, there is some wierd stuff going on, i need to figure out this new auth scheme
[08:02:54] <navneet894> Hello, Iam Navneet Mittal from IIT JODHPUR , i know c++ and want to contribute to mongodb. Iam new to open source and need guidance.
[08:16:02] <navneet894> i just want to give a start to open source contribution? want to know what i can contribute?
[12:16:41] <mn3monic> hello, about db.collection.insert({'sample_key': 'sample_value'}), how I specify that 'sample_key' must be unique without quering at every single insertion to check if it's already in the db?
[12:20:42] <mn3monic> if not db.collection.find_one({'sample_key': 'sample_value'}): db.collection.insert({'sample_key': 'sample_value', 'second_key': 'second_value', ...})
[12:23:08] <brammator> from collection {name, id, lastupdate}, and list of names plus timestamp. Could I have three results: "names not in collection; names in collection but lastupdate < timestamp; names not in collection at all" in one request?
[15:11:16] <brammator> I have to cache some API requests. What's better: use compound key{host, path, params} or simple key with (host,key,params) converted to string?
[15:47:07] <DarkLinkXXXX> Can mongoimport work with just any json file, or does it require something mongo-specific?
[15:48:31] <kali> DarkLinkXXXX: by default, it expects one json object exactly per line, and will be happy to load it whatever the content is
[15:49:41] <kali> DarkLinkXXXX: that said, do not expect miracles for non-json types like dates and binary
[15:50:45] <DeveloperDude> Hello everyone! I have a question that is probably very common, but I've found several answers and I'm not sure what is the proper way of doing this.
[15:50:53] <DeveloperDude> The thing is that I have a users collections and an events collection, coming from a relational background I was storing userids inside the events collection, but there are no joins in mongodb.
[15:51:53] <kali> DeveloperDude: have a look at that: http://blog.mongodb.org/post/65612078649/schema-design-for-social-inboxes-in-mongodb
[15:52:58] <DeveloperDude> I supposed that, it seems dirty, but its mentioned in some blogs and forums
[17:24:16] <whomp> how can i speed up my mongoimports? for example, i was thinking maybe i could convert the json to bson and import it or something
[22:00:35] <sudormf> Question. If I have 20 tables in a nosql database, and i have one database node, but its getting overloaded, so i add 3 more database nodes. now i have 4 total db nodes. will my db scale automatically to use these 4 nodes, or will i have to do any manual configuration to be able to use them?
[22:04:06] <jumpman> hey, i have a quick best practice question. i'm going for performance in a reasonably large database
[22:06:33] <jumpman> ...anyways, i'm trying to store a 2d plane of 'solar systems' each with 'planets' and each planet with a 'puzzle'. but i want to have ~15,000 solar systems with 5-10 planets each.
[22:06:55] <jumpman> i'm planning on storing the 2d plane in chunks - each with a 'solar system' location to be drawn and a position and id inside
[22:07:38] <jumpman> then the id found in a separate collection of each 'solar system' which would include the id, and planets
[22:08:13] <jumpman> here's where the question comes in: would it be faster to store all level data in one collection with 'puzzle id' so that the 'solar systems' call is smaller
[22:08:30] <jumpman> or better to store the puzzles in the planets and just have larger units in starsystems
[22:10:06] <jumpman> basically i'm trying to decide between one collection containing ~1,000,000 puzzles and one collection containing all of the ~15,000 solar systems with the same data from the first collection
[22:12:08] <jumpman> right, that's what i figured. i just didn't know if the amount of data within the entries would have the same impact
[22:21:25] <tornado_terran> im not familiar with mongodb specifics, what is better to use aggregation framework or to fetch a lot of rows with few fields
[22:22:22] <tornado_terran> i would like to get avarage/median value. I can use aggregation framework or in my case just fetch all rows from some time bucket with fields like firstActionAt, lastActionAT
[22:24:08] <tornado_terran> im trying to create analytic tool based on golang and mongodb. im not familiar with mongodb specifics, what is better to use aggregation framework or to fetch a lot of rows with few fields. i would like to get avarage/median value. I can use aggregation framework or in my case just fetch all rows from some time bucket with fields like firstActionAt, lastActionAT
[22:46:37] <sputnik13_> I'm trying to add a user to a database remotely using pymongo and it just fails silently
[22:47:21] <sputnik13_> meanwhile if I add the user through the mongo cli client to the database it succeeds
[23:02:50] <sputnik13_> deanclkclk: http://stackoverflow.com/questions/6697871/transfer-mongodb-to-another-server