Log file Viewer

[01:07:50] <rishabhm> im new around, do we just ask our mongo related questions here?

[01:08:33] <rishabhm> is anyone around?

[01:53:25] <joannac> rishabhm: yes, just ask your question

[03:07:02] <jason0_> Quick question about the nodejs mognodb interface. If you have a cursor how do you grab the next batch?

[03:38:41] <gancl> ehershey: Your are right. Thanks

[05:06:32] <QubeZ> hi all

[07:54:05] <solarradiation> does mongodb's aggregation framework support filtering data through arbitrary external applications?

[08:46:11] <kali> solarradiation: nope

[08:51:22] <solarradiation> kali that kinda sucks

[15:25:21] <FingerPuk> Afternoon all

[15:26:38] <FingerPuk> It may be that I’m tired but I’m struggling to think of the best way to structure a very simple thing. Making a tea rounds app, wondering if I should do it by: People collection, person is in a round OR Rounds collection that contains people. I’m building it to learn Meteor.

[15:27:21] <FingerPuk> Thinking that Meteor has a users collection by default so I could use that, but if it was structured by rounds it would be simpler down the road, unless people want to change. Maybe I need to sleep.

[15:36:34] <Gargoyle> FingerPuk: Make a brew and have a think about it! :P

[15:37:26] <FingerPuk> I’m going to do it by person and see how that works out :)

[15:38:24] <Gargoyle> I'd probably store the drink preferences in the people collection. (Tea, Milk, 1 Sugar)

[15:38:41] <Gargoyle> Then just have lists of people in the rounds collection.

[18:13:07] <c3l> In a collection I have lots of documents which have tags: [String

[18:13:53] <c3l> ], I want to find all existing tags. How do I best do this? Ive heard about map-reduce and the aggregation pipeline, are any of those tools well suited for the job?

[18:16:32] <c3l> Or do you think that it is bad to store tags like this? Should I create a collection for tags and keep references instead of plain strings for the tags? They are to be created and changed by user at any time.

[18:56:09] <dani_> hello

[18:56:30] <dani_> How do I delete a set of objects I queried in some way with mongoalchemy?

[19:33:57] <neutralizer> hi, any idea why I am getting this? 06-09 00:32:56.119: E/AndroidRuntime(5813): java.lang.NoClassDefFoundError: com.mongodb.MongoClientURI

[19:34:33] <neutralizer> app is compiling fine and this is runtime error

[19:36:54] <neutralizer> or do you have any example for mongodb java driver version 2.12.2

[19:38:37] <neutralizer> here's the code http://paste.ubuntu.com/7614556/

[19:44:23] <cheeser> you need to include that jar in your classpath

[19:49:11] <neutralizer> cheeser: it is there

[19:49:15] <neutralizer> already

[19:49:21] <neutralizer> that's how it's compiling fine

[19:49:40] <cheeser> compiling and runtime both have their own classpaths.

[19:53:01] <neutralizer> cheeser: so I am using eclipse

[19:53:45] <cheeser> ... ok

[19:54:19] <neutralizer> so I mean to ask if isn't it as easy as adding jar to build path?

[19:54:25] <neutralizer> won't it set for both?

[19:55:21] <cheeser> i don't use eclipse but i'd expect so, yes. are you seeing this error in eclipse or on your android device?

[19:55:37] <neutralizer> Eclipse LogCat

[19:55:41] <cheeser> in the build settings see if there's a scope (compile vs runtime) you can play with

[19:57:07] <neutralizer> I can confirm this. It used to work some time ago with different eclipse version.

[19:57:39] <neutralizer> I have used mongodb with JS and Python and tried to test Java driver

[19:58:00] <cheeser> yeah, i can't really help you with eclipse.

[19:58:27] <neutralizer> no problem, thanks for that much

[20:05:15] <neutralizer> cheeser: removed jar and re-added and now it is working :S

[20:05:24] <neutralizer> but with different error :)

[20:05:45] <neutralizer> 06-09 01:04:42.139: E/dalvikvm(6288): Could not find class 'javax.management.ObjectName', referenced from method com.mongodb.util.management.jmx.JMXMBeanServer.createObjectName

[21:52:05] <brom> hi, is a collection - document and does it have 16MB limit?

[22:20:39] <jason0_> When using a mongo cursor how do you get the next batch of elements?

[22:22:58] <cheeser> it does that automatically

[22:25:58] <jason0_> cheeser: from what I can tell it stops returning after 100 elements.

[22:26:40] <cheeser> what are you using to query?

[22:27:14] <jason0_> cheeser: querying from a simple nodejs app using the mongo driver

[22:27:47] <jason0_> cheeser: as a test I supplied .limit(1000) and the cursor did indeed return more elements.

[22:28:01] <cheeser> make sure the app isn't doing that. the driver itself won't impose that limit.

[22:28:26] <jason0_> cheeser: yeah I'm not invoking with al imit

[23:26:55] <usr> Is there any cons in using python to make and maintain mongo databases?

[23:46:40] <JamesHarrison> quick question - just upgraded from 2.4.10 to 2.6.1, now seeing queries that used to take ~20ms take ~1000ms; are of the form { field_with_index: false, _id: { $in: [ list of objectids ] } }. These are small result sets (~20 docs) on a smallish collection (~600k docs). planSummary is IXSCAN {_id: 1}.

[23:47:04] <JamesHarrison> Any idea why the runtime would be so much higher post-upgrade/what I should be looking into, metrics-wise, to figure out the cause of the slow performance?

[23:49:13] <JamesHarrison> (workload is read-heavy, ~1k queries/sec, ~100 insert/updates per sec, database lock % varies between 0.5 (mostly) with occasional very high peaks of ~60% once every 15-25 seconds which concerns me)