[00:35:52] <gsd> if im connecting to multiple hosts, do i need to prepend user@pass: to each host in the connection string
[00:45:47] <sheki> hey i have a query which looks like { a : "some string", b : {$gt : new Date(1393229650673)}} =>
[00:45:52] <sheki> i am looking to add an compound index
[00:45:57] <sheki> what should the compound index be.
[00:46:17] <sheki> from what i know the index should be { a : 1, b :1 },
[00:46:33] <sheki> does the order on field b matter?
[01:57:59] <ben2> hey all, I have a collection of users with many fields I need to filter by. I want to select data from a log collection that belongs only to the subset of users I filtered by. Some filters can return 100k users which seems like a lot of data to send back and forth if I were to first select all the user_ids (all 100k) and use that to filter log data. Is there a better way to handle this?
[02:47:19] <the8thbit> I'm having trouble connecting to my mongodb on an openshift account. When I ssh in and try to enter the mongo shell I get:
[02:47:20] <the8thbit> Thu Mar 6 21:32:01.272 Error: 18 { code: 18, ok: 0.0, errmsg: "auth fails" } at src/mongo/shell/db.js:228
[03:13:22] <the8thbit> joannac: I'm also getting an error any time I try to do a get query, so if all I need to do is supply credentials there then I dont think the issues are related
[03:35:19] <the8thbit> remotely I have db users, but I don't have the users db that I'm using to manage users... which is why I think I'm getting that error when I try to log in
[05:40:40] <brucelee> for mongodb + ssl, can we simply compile the source to install it, and then the just save the produced folder to subsequent servers
[05:41:01] <brucelee> or do we have to compile it over again for the next mongodb installation?
[06:41:35] <scruz> hello. been trying to do this https://dpaste.de/o6dw with the aggregation framework. any help, please?
[07:20:56] <scruz> kali: just found http://www.kamsky.org/1/post/2013/04/using-aggregation-framework-to-reshape-schema.html
[07:28:39] <kali> yes, indeed, but you need to know all the possible keys beforehand
[11:06:24] <danpalmer> I'm doing a research project at uni, analysing MongoDB in terms of distributed systems, I was wondering if some people here could give me some input.
[11:07:30] <danpalmer> I'm trying to figure out Replica Sets and leader elections stuff. There doesn't seem to be an algorithm like Paxos/Raft/etc behind it, and I can't get much detail on how things like partial network failures and latency are handled.
[11:08:46] <danpalmer> Is it the desire for the sort of 'tunable' consistency in MongoDB that decided that? Those consensus algorithms are typically used for distributed transactions.
[12:18:44] <GHJGH> afternoon all, how do you go about naming your mongo servers? ive come up with - {env}-{service}-mdb-{replica-set}-s{#shard}-{#server}, is there a better way?
[12:21:18] <kees_> i gave them names from greek mythology and remember what it does, but if you want to use those names, don't user abbr too much
[12:30:45] <GHJGH> greeks cool, but i need something a little more 'webscale'
[13:22:03] <NoReflex> hello! Can i recreate a binary object from the it's string representation?
[13:23:03] <NoReflex> I only have a "paste" of the actual binary data from a logfile; I cannot access the mongodb instance where the actual binary data is stored
[15:25:08] <obihann> I know this isn't the mongoose channel but their channel is dead
[15:25:12] <obihann> with the following model, how would I perform a find a populate on the members.user object? https://gist.github.com/anonymous/9413322
[16:43:35] <freeMind> Hi every body, im using mongoDB with c++ driver i want to get list of filed names of each document in my collection if u can help me
[17:03:39] <codydh> Hello! Silly question, but if I apt-get install mongodb on Ubuntu, how do I figure out the necessary user/pass to connect using a local Node.JS app?
[17:25:20] <chandru_in> If the page of index to be updated is not in memory, will mongod yield the lock while fetching the page?
[17:25:52] <nukulb> joannac: quick question for you when you get a minute. on the text search docs - http://docs.mongodb.org/manual/tutorial/enable-text-search/ it say its Beta feature and don't enable it on Production. What are the consequences of using this in Production?
[18:31:20] <joannac> nukulb: the same as any other beta feature? it may not work, it may have bugs, it may have effects you don't expect, the implementation details (e.g. API) might change
[18:32:43] <nukulb> joannac: is there a timeline for something like this to come out of Beta and become a full feature? doesn't have to be accurate just looking at it so I can plan better to switch over to this
[18:36:03] <nukulb> joannac: more importantly I am looking for a way to do a weight search other than using a full text search. So basically be able to do a search using $regex with different properties get different weights and then sort by that . Do you know of any examples?
[18:38:43] <tscanausa> do replicate sets allow for ip addresses for naming members?
[19:34:48] <avelino> hi guys, 'm working in open source BI application, per default used riak, the monongodb team want to sponsor to implementation mongodb alternative data base?
[20:29:40] <saml> should i get an interview and make mongodb better?
[20:29:51] <tscanausa> joannac: looks like cssh was messing with mongo shell
[20:30:38] <joannac> tscanausa: also, you're on a very old version of mongodb which means you may be hitting a bug I'm not familiar with
[20:33:00] <tscanausa> I just noticed the old version, Ill change that
[20:33:11] <mboman> I was thinking of plotting some values from my mongodb collection, what would be the best way to do this? Own python program, or has something already been created?
[21:54:34] <harttho> Say you want to store X documents a day (where X > 10mm and is distributed over the day), the order of the documents doesn't matter, but querying them on time_entered (separate than _id) is important
[21:54:58] <harttho> There will also be a few indexes on other fields in the document
[21:55:17] <harttho> How do you keep the indexes getting so large they stop being nearly as useful?
[21:56:43] <harttho> Originally we had a separate collections by day, but querying can get annoying having them separated
[22:12:33] <synth_> hello, i'm curious what the best method for checking if a record in a collection contains an attribute?
[22:14:21] <synth_> okay, so in python, i tried doing this, however i'm not sure if i was doing it correctly.... here's my code: self._user_collection.find({"uid":uid}, {attr:{"$exists":True}}); where "attr" is an attribute like "phone"... using that code, i was still able to return a cursor even though the attr didn't exist
[22:15:11] <cheeser> was there anything in the cursor?
[22:16:18] <synth_> see, that's where i was a bit confused. i thought it would return None
[22:32:59] <iksik_> hello, it's my first time with mongorestore https://gist.github.com/krzysztofantczak/6e8b07223abee3743356 - but there is one weird thing, api collection is empty after restore, any ideas why?
[22:34:35] <cheeser> what's with that "db" option there?
[22:37:00] <iksik_> i was also trying to copy this one collection using robomongo tool with exactly same result, all collections are fine but 'api' is empty :|
[22:38:47] <iksik_> example document from 'api' collection looks like this https://gist.github.com/krzysztofantczak/9814af36141d1cb2812f - if it matters
[22:39:46] <cheeser> connect to mongo with no params, then "show dbs"
[22:41:29] <iksik_> cheeser to my target database? where i'm trying to restore that collection?