Log file Viewer
#mongodb logs for Friday the 8th of March, 2013
[00:19:29] <FrankBullitt> Is there a tool like mongoimport that will follow/tail a log and continually import JSON?
[00:21:30] <stdim> Hi. How fast are array queries for _id. For example: findOne({ 'parentArray._id': ID }) ?
[00:48:08] <bootstrapper> db.items.aggregate( { $group: { "_id": { id: "$id" } , version: { $max: '$version' } } , $sort: {'version' : -1 }} ).result
[00:50:09] <FrankBullitt> Is there a tool like mongoimport that will follow/tail a log and continually import JSON?
[00:59:38] <dstanek> the data in my dbpath totals about 29G while my backup is only about 7G - why such a big difference?
[01:01:09] <bootstrapper> > db.items.aggregate( { $group: { "_id": { id: "$id" } , version: { $max: '$version' } } , $sort: {'version' : -1 }} ).result.length
[01:01:10] <bootstrapper> > db.items.aggregate( { $group: { "_id": { id: "$id" } , version: { $max: '$version' } } } ).result.length
[01:01:45] <bootstrapper> i want to get the latests objects, where each object is the latest version
[01:05:12] <stdim> Returns the highest value among all values of the field in all documents selected by this group.
[01:08:36] <stdim> Please do, since I don't know how MongoDB would know what is the 'highest version' of a document...
[01:08:57] <stdim> Your documents don't seem to have references to past versions or a common grouping element...
[01:16:08] <stdim> db.items.aggregate( { $group: { "_id: "$id", "version": { $max: "$version" } } })
[02:39:31] <owen1> after restoring my data i noticed that the size of my db is big, but i don't see any records. any ideas?
[02:50:53] <ferai> does anyone know why using .sort({"$natural": -1}) would fail to return any documents, but .sort({"$natural": 1}) would return all documents?
[03:44:29] <frozenlock> Hello gentlemen, I'm trying to add indexes, but can't seem to get mongo to use them... is there a minimum number of documents to get before the 'cursor' switches to BTREE?
[04:19:19] <zarac> Reading the documentation for the node.js driver; ( http://mongodb.github.com/node-mongodb-native/api-generated/collection.html#find ) it's stated that if i give find(..) a callback, the second parameter will contain the result. What kind of an object is this?
[04:20:39] <zarac> It looks very similar to what gets returned (a cursor). I'd just like to get an array (without having to use the toArray method like the examples do).
[04:32:52] <zarac> Hmm.. it seems the documentation inconsistent with the driver. Anyone else got an opinion on this?
[04:49:14] <LuckyBurger> hmm. After executing a loop of db.collection.save(), how do I tell node that all db ops are done and to disconnect?
[04:59:05] <zarac> Dunno.. anyway. One solution is to have a counter in your callback for .save and when it hits 0 you call your cleanUp() function.
[04:59:56] <zarac> I'm fairly new to javascript / node.js / mongodb myself.. i bet there's a better solution.
[05:03:28] <LuckyBurger> which means its time for me to goto bed cuz im being an idoit on basic things rofl
[05:05:26] <zarac> LuckyBurger: http://stackoverflow.com/questions/728360/copying-an-object-in-javascript might enlighten you (and me)
[06:24:34] <shmoon> i have a users collection, i want to generate small unique ids for each (they dont have usernames, so ...). my initial thought is to implement auto increment ids, how do i go about doing that, and is that sane?
[09:26:41] <mtsr> Hi! If 'db.questions.find({ 'user.email': 'bla@example.com' })' returns multiple documents, shouldn't 'db.questions.update({ 'user.email': 'bla@example.com' }, { $set: { 'user.test': 'bla' }})' also update multiple documents? Since that's not what I'm seeing.
[09:27:51] <Gargoyle> mtsr: You have to explicitly specify that you are intending to update multiple docs with one query.
[10:11:25] <Nodex> on a slightly off topic note, this whole responsive website malarkey is easier than people make out
[10:24:13] <remonvv> You have to be permanently online to play, you can't log in about 60% of the time because they can't even get a scalable auth server right.
[10:24:58] <Zelest> expensive.. fair enough.. but complaining about being online? how often are you offline?
[10:25:18] <remonvv> Zelest, I'm not complaining about it. That's actually the only part of the shitstorm I don't agree with.
[10:25:25] <Nodex> Zelest : from what I gather... when the servers go down, your city can't make money and you lose
[10:25:41] <remonvv> The problem is that IF you want to require online online your servers have to work.
[10:26:12] <remonvv> So you're stuck with a game that doesn't work because it requires you to be online for a DRM check rather than for gameplay.
[10:27:44] <remonvv> It's the movie discussion all over again. Studios release movies here 6 months later and then bitch people pirate them.
[10:27:58] <remonvv> They don't pirate because they're cheap bastards usually, it's because they don't want to wait 6 months for the bloody thing.
[10:31:02] <remonvv> Yeah but those only because foreign broadcasters wanted to pay huge sums for multi-year licenses
[10:31:36] <remonvv> EA isn't run by gaming people. I've worked for them and noticed that more than anything.
[10:32:29] <remonvv> Ignoring the fact that just making great shit and getting more consumers is as effective, if not more so from a brand reputation point of view
[10:32:59] <remonvv> Anyway, this is sideline commenting, I'm sure there are a lot of factors involved.
[12:53:18] <rickibalboa> Would there be a way to keep an array inside a document at a fixed size, ie $pop'ing when an element is $push'ed and the array is of a certain size without any application logic, just in queries?
[14:33:42] <richwol> Does anyone know if it's possible to use the aggregation framework to perform a group on a subdocument (not grouping documents together)
[14:38:58] <fishfish> hi everyone! does anyone know where does mongo store its log files by default while in verbose mode?
[14:49:55] <alex88> hello guys, i'm new to mongodb, actually I've got how to find with a simple filter, suppose that I've an element like { metrics: [["key", 123],[...]] }
[14:59:01] <alex88> one last question, I've actually var metrics = collection.find({}, { metrics: 1, internalHostname: 1, collection_timestamp: 1 }).toArray(function(err, data) { ... }
[14:59:44] <alex88> if I do var singleMetric = collection.findOne({}, { metrics: 1, internalHostname: 1, collection_timestamp: 1 });
[15:14:49] <Derick> $MONGO["servers"][$i]["control_auth"] = false;//enable control users, works only if mongo_auth=false
[15:22:58] <alex88> Derick: nope? http://docs.mongodb.org/manual/reference/method/db.collection.findOne/#db.collection.findOne isn't this one?
[15:25:36] <Derick> yes, you referenced the *shell* docs, which are also in JS , but not the same as node
[15:27:08] <Derick> alex88: you need to refer to: http://mongodb.github.com/node-mongodb-native/api-generated/collection.html#findone
[15:35:17] <fommil> hi all - if I use stdout for mongodump, is it just the BSON that is piped? is the JSON metadata simply ignored?
[16:18:38] <nonent> i've been using the $inc operator to do realtime statistics for quite a while with mongo. we're pushing a lot of io, and i'm curious if a) every individual $inc gets written to journal, and likewise if each $inc has to be propagated individually over oplog replication.
[16:20:07] <nonent> i'm thinking about locally caching the stats in app-space instead of inc'ing each individually, and wondering if it's worth the effort (and loss of true realtime), or if the way mongo works at a deep level means i won't actually see efficiency gained
[16:36:33] <jiffe98> I removed the content on one of the config servers it was complaining about and restarted it and it rebuilt but I'm still getting complaints that it differs
[16:44:03] <lethjakman> hey, does anyone have a good resource on learning how to query and insert mongo?
[16:44:11] <kali> jiffe98: http://docs.mongodb.org/manual/tutorial/manage-sharded-cluster-config-server/#sharding-config-server-replace
[16:53:33] <bmcgee> hey guys, are any of you familiar with a bitemporal data model? Better still, implemented one with mongo? (I know it's a longshot)
[17:19:26] <bmcgee> hey guys, having trouble devising an aggregation. Think i'm most of the way: https://gist.github.com/brianmcgee/5118104
[17:29:35] <JakePee> what are the differences between using a boolean vs an integer in relation to indexes and querying
[17:38:58] <bmcgee> Derick: I've elaborated a bit on my problem in the gist: https://gist.github.com/brianmcgee/5118104 taking into account your suggestion. Am i using $first correctly?
[17:41:00] <bmcgee> Derick: don't think i'm using first correctly, i don't want to pick out values, I want to grab the first entry for each grouping
[19:08:43] <bmcgee> when grouping is it possible to specify the first instance of the grouping for each _id
[19:11:28] <bmcgee> i've looked at first but i don't want to extract the first value of a field, i want to extract the first document for the grouping
[20:55:30] <templaedhel> Hello, I have a collection with documents that looks like this https://gist.github.com/anonymous/16e94e3cad3c0c111ace and I want to do a query to find all the documents that "cover" a point. Aka a point exists within the bounds. What would the best practice be for something like that?
[23:37:11] <bobinator60> micromongo requires all models to be in their own collection, and mongoengine is just way too big