Log file Viewer
#mongodb logs for Monday the 27th of May, 2013
[06:07:53] <venturecommunist> in javascript i might say db.somecollection.find({attribute: "attribute"}) which works fine but what i'd really like is db.somecollection.find({attribute: "attribute"}).numberofhitsigot
[06:08:32] <venturecommunist> do i have to index and explain (something i'm just learning about) or is there a simpler way
[06:15:40] <sb8244> db.test.find( { loc : { $near : { $geometry : { type : "Point" , coordinates : [ 126.93521976470947 , 36.80358081325186 ] } , $maxDistance : 1000 } } } )
[06:18:38] <sb8244> okay so going with my last message, can anyone help me figure out what the hell is going on? I can't see what I'm doing wrong
[06:58:48] <sb8244> oh I fixed it! Was on 2.0.4 but needed to be on 2.4 for 2dsphere indexes / queries
[09:16:30] <armetiz> But I guess the problem come from nginx-gridfs which doesn't cache chunks and compute request and chunck compilation
[09:17:29] <armetiz> And before patch nginx-gridfs with a cache system using md5 of fs.files, I wondering if nginx -> gridFS FUSE -> gridFS can't be a good solution
[09:39:02] <maasaki> i used mongoose with Node JS. I want to have sub documents inside the parent document. I want to have child document as separate collection. I declared like 'children: [childSchema]' inside parent schema. then i save it like var parent = new Parent({ children: [{ name: 'Matt' }, { name: 'Sarah' }] })
[09:40:00] <maasaki> but child documents are not stored in the sperate collection. It is stored inside the parent documentn itselft. Why it that happening?
[09:41:46] <jareiko> I'm making a high score table, and I'd like to get the best N scores, but with each player only appearing at most once
[10:15:53] <guest1924> Do I need to perform any action after I update my collection with Mongo pecl driver?
[10:16:22] <guest1924> it keeps on returning info DFM::findAll(): extent 0:2000 was empty, skipping ahead.
[11:12:22] <maasaki> how to store sub document in a separate collection while saving the parent collection?
[12:29:09] <EmmanuelJacobs> Hi guys, trying to understand a strange behavior using mongodb 2.4.3 win32.
[12:29:42] <EmmanuelJacobs> I'm inserting documents using a stored function as one of the parameter and it seems that function is called several times at each insertion.
[12:30:18] <EmmanuelJacobs> I have a counter initialized like this: db.counters.insert( { _id: "uqid", seq: NumberLong(0) } );
[12:31:17] <EmmanuelJacobs> I have a stored function named getUqid which is function () { var ret = db.counters.findAndModify( { query: { _id: "uqid" }, update: { $inc: { seq: NumberLong(1) } }, new: true } ); return ret.seq; }
[12:31:36] <EmmanuelJacobs> When I do three insertion like this: $conn->test->ads->insert(['qid' => new MongoCode('getUqid()') , 'name' => "Sarah C."]);
[12:31:56] <EmmanuelJacobs> { "_id" : ObjectId("51a34f8bf0774cac03000000"), "qid" : 17, "name" : "Sarah C." }
[12:31:56] <EmmanuelJacobs> { "_id" : ObjectId("51a34f8bf0774cac03000001"), "qid" : 20, "name" : "Michel D." }
[12:31:56] <EmmanuelJacobs> { "_id" : ObjectId("51a34f8bf0774cac03000002"), "qid" : 23, "name" : "Robert U." }
[13:46:44] <patcito> if I have the value of an _id, is it possible to get the latest 5 documents in a collection starting from that _id?
[13:56:53] <patcito> figured it out (using mongoid) File.where(:_id=>{:$lte=>doc._id}).sort(_id:-1).limit(5)
[14:28:42] <failshell> hello. when i disable the balancer on a config node and then run sh.getBalancerState() on a replica set node, it should return false right?
[14:28:50] <kobigurk> regarding Azure on MongoDB - I want to use the RoleEnvironment.Changed event and reconfigure mongo as to the number of replica members existing there. It's relativly simply - is there any reason it wasn't implemented? Am I missing something?
[14:34:24] <baniir> anyone running raid10 ebs setup with ubuntu? I set up according to the docs at http://docs.mongodb.org/ecosystem/tutorial/install-mongodb-on-amazon-ec2/ but am getting readahead warnings from mongo; i think the readahead settings are not persisting through reboots
[14:36:25] <failshell> hello. when i disable the balancer on a config node and then run sh.getBalancerState() on a replica set node, it should return false right?
[14:50:19] <ExxKA> Hey Guys, I have a question on mongoose that I hope some of you can answer. When I build up an object graph, I later experience that at a certain depth, only the object id's are returned, not the actual object. Why is that?
[16:13:19] <OliverJAsh> I want to build an application similar to Twitter, and I'm looking for advice/help/recommendations on how to engineer the back-end (in terms of collections, replication, etc.)
[16:14:09] <OliverJAsh> Currently the `User` document has an array on it called `followingUsers`. When I want to load that user's timeline, I look for `Tweet`s that contain the user IDs in that array/
[16:25:25] <double_p> i hate the connection drops for rs.reconfigure() :-( is that contained within shards aswell?
[16:31:34] <double_p> i know when it happens, but i didnt read much about sharding so far. just curious..
[16:31:58] <double_p> the app servers really dont like that "only some (micro)seconds" loss .. or let's say, customer does not :)
[16:34:39] <double_p> just the point, that a backup/DR-site slave is sometimes disconnected (vpn drop). and even with a replSet of main/local-slave/dr-slave with a priority 3/2/1 they always start to vote when dr-slave goes dark. and i want to contain that
[18:07:08] <xuewenzhi> Hi, I have a feed system using fan-out on write. I keep a list of feed ids in redis, and save the feed content in mongodb, so every time when i read 30 feed, i have to do 30 query to mongodb, is there anyway to improve it ?
[18:12:00] <redfox_> hi all. Im having a document and want to remove a subdocuments object. Is an update({}, $pull: { _id: ... }, { multi: true }) as performant as an update({ _id: ... }, { $pull: { subdoc: { _id: ... } } } ) ?
[18:15:13] <redfox_> in other words: how does it affect performance to $pull a subdocument with searching through the whole collection, of with selecting the parent document first?
[18:17:29] <tncardoso> I am experiencing high memory usage in my server. I have a machine with 17GB RAM, and it is using 14GB resident and 60GB virtual. My main concern is the non-mapped virtual memory usage of 14GB. All my databases and collections use less than 8GB. What are the possible causes?
[18:22:44] <tncardoso> double_p: kali thanks for the answer. I've already checked this link. The problem is that all my indexes and all the storage have 8GB. In theory I could fit all my working set in memory.
[18:23:54] <double_p> tncardoso: pick it that way. mongo is a mem-hogger (for virtual). we upgraded from 8G to 96G. and mongo wasnt shy to allocate that
[18:27:13] <double_p> the point in case, 6 fold the mem. mongo goes there... "spread" -- but speed? well :D
[18:37:49] <kali> tncardoso: i don't know where these 14GB comes. i'm pretty sure it's not the 100 connections
[18:39:40] <tncardoso> once i had problems with gentoo (the emerge tree got completely broken). bad times
[18:59:25] <baniir> hi. is anyone here running raid10 on ebs? i changed readahead values with blockdev, but i don't think the settings are persisting through reboots
[19:16:04] <ixti> I clearly understand that JS does not distinct Float vs Integer... it's all Number()
[22:08:52] <arthurnn> when I am draining a shard. the reads from the draining happen on a secondary or in the primary? I would assume it is on the primary so it moves the more consistent data. is that right?
[23:41:56] <BurtyB> meltz bit quiet in here atm but people might come to life if they know the Q otherwise hang around ;)
[23:42:47] <meltz> my question is what is the best way to do a filesystem level backup on a replica set configuration where each database is on a RAID 10 config