Log file Viewer

[00:03:54] <dimon222> sententiousness, stackoverflow

[00:04:07] <dimon222> alexi5, not really

[00:04:39] <sententiousness> thx, was hoping for an irc channel

[00:05:54] <dimon222> well, its hard to compare irc with few hundred ppl to several thousands of ppl on stackoverflow network

[00:06:23] <dimon222> more like hundred of thousands, but well

[00:09:59] <alexi5> i did a test by import a 1.5GB csv file into mongodb and supprise to see the mongodb database grew to 9.3GB

[00:10:35] <alexi5> i guess i did something wrong

[00:26:42] <dimon222> hoh, thats definitely strange

[00:27:08] <dimon222> did you have indexes set up?

[00:29:35] <alexi5> no indexes setup

[00:29:58] <dimon222> hm, did you check actual collection?

[00:30:37] <dimon222> in my opinion it shouldn't really be more than 2.5 GB (because of _id field)

[00:31:20] <dimon222> preallocation is set up by default to 2GB, so cant really be more than 4GB, 9.3GB is something very very strange

[00:32:34] <alexi5> the csv file contain over 7 million rows

[00:33:02] <alexi5> and each row has about 15 fields

[00:33:18] <alexi5> would that be the reason why ?

[00:34:05] <dimon222> only if it creates indexes, but it doesn't seem so

[00:34:11] <dimon222> because indexes really can be heavy

[00:35:19] <alexi5> i'll drop the database and re import and see if there is any difference

[00:38:23] <dimon222> incase there's no difference, try to check deeply structure in mongo

[00:38:32] <dimon222> like call dbStats on database through shell

[00:38:34] <dimon222> http://docs.mongodb.org/manual/reference/method/db.stats/#db.stats

[00:38:46] <alexi5> ok

[00:46:05] <alexi5> hi dimon222

[00:46:15] <alexi5> here is the stats for the database : http://pastie.org/9824808

[01:07:33] <dimon222> strange, maybe stuff like keynames doubles the size

[01:07:52] <dimon222> so in csv you have header line, but in collection its actual set of pairs for each document

[04:18:25] <anybroad> hi

[04:18:28] <anybroad> Where is this list function actually used?: https://github.com/madhums/node-express-mongoose-demo/blob/master/app/models/article.js#L179

[04:18:31] <anybroad> I cannot find the place where it is used.

[05:51:12] <tommyvn> morning all, I've run into this exact error on my mongo enterprise installation and was told this is the place to get some help fixinig it: https://jira.mongodb.org/browse/SERVER-15369

[05:51:32] <tommyvn> the error reads "Assertion failure _name == nsToDatabaseSubstring( ns ) src/mongo/db/catalog/database.cpp 472"

[05:55:48] <joannac> tommyvn: if you are using enterprise, don't you have a subscription?

[05:56:19] <JakeTheAfroPedob> hi guys anyone with json experience here?

[05:56:28] <joannac> also, where were you told that this is the place to get some help fixing it?

[05:57:29] <tommyvn> I'd imagine so but http://www.mongodb.org/about/support/ said to head over here. maybe I need to dig around the mongo site a little more? I'll check for more info in the links on that page

[05:58:07] <joannac> tommyvn: if you have a subscription you should be opening a ticket, not asking irc

[05:59:16] <joannac> in any case you need to fix up your files. is it production?

[05:59:49] <joannac> JakeTheAfroPedob: i have no idea what that means; you should probably just ask your question

[06:00:51] <tommyvn> thanks Joannac, I didn't realise, that mongo support page might save you guys a few questions if it said exactly that.

[06:00:58] <tommyvn> it is indeed in production

[06:01:32] <JakeTheAfroPedob> jonnac: i have a json file. It's like XML. Formatted data. Is there any way i can write a program to split all the clustered datas into its respective spereated files

[06:02:10] <joannac> tommyvn: that page is for community users. When you pay for a subscription you should've gotten links for how to take advantage of the support subscription :)

[06:02:37] <joannac> JakeTheAfroPedob: sure

[06:02:51] <joannac> tommyvn: replset? shard? how many nodes affected?

[06:02:55] <JakeTheAfroPedob> jonnac: can i do it with for loops only?

[06:02:56] <tommyvn> right, we probably did but there's loads of people on my team so someone else probably put them on our wiki somewhere and didn't mention that's the path to take

[06:02:59] <joannac> do you have backups?

[06:03:12] <tommyvn> checking....

[06:03:53] <joannac> tommyvn: PM me the name of your company?

[06:17:08] <tommyvn> joannac: to answer your previous questions, 4 node replset, no sharding, so far one but only because the host locked up and had to be rebooted, I don't know if I'd see this issue on running nodes or not

[06:17:36] <joannac> tommyvn: I've PMed you

[06:17:36] <tommyvn> I hit the issue on node startup, the other 3 are still running

[06:19:50] <JakeTheAfroPedob> jonnac: any idea? i got dc'ed

[06:43:53] <joannac> JakeTheAfroPedob: i have no doubt it's possible to parse a JSON file however you like. This channel is for help with MongoDB though, you should ask in a programming forum

[06:54:16] <JakeTheAfroPedob> jonaac: thanks

[10:15:59] <brano543> Hello, i would like to use mongodb for spatial search and text search based on the name of the street. Can i use mongodb for this purpose or do i need create R-tree for this kind of job? It may help my application if all neighbours of particular node would be in the same bucket,it is easier for me to find local POIs.

[17:34:40] <anybroad> hi

[17:34:44] <anybroad> So I use mongoose.

[17:35:01] <anybroad> Well, what do I have to do when I want to change the schema? Will mongoose update the existing schema in mongodb?

[17:38:18] <anybroad> anyone?

[17:58:00] <disorder20> hello

[17:58:18] <disorder20> I have an array of filters to put in a $or statement

[17:58:25] <disorder20> i.e. phone, name, surname ecc

[17:58:47] <disorder20> when a field is undefined, I want the query to find all documents

[17:58:54] <disorder20> is there an easy way to do this?

[17:59:06] <disorder20> (I'm using node.js as client)

[18:51:37] <wieshka> Hi ... having problem that for my replicaSet with 3 members on all servers free disk space 0% .... recovery plan in mind: 1) shut down node3 to have full backup, 2) empty database from old entries freeing up a lot of space on node1 .... 3) remove all data directory on node2, wait till synchronizes with node1. 4) delete all data on node1, wait till synchornizes with node2 .... 5) if all good, delete all data from node3 ... 6) syncing all nodes back .....

[18:51:37] <wieshka> expected result: all nodes synced with released space ....... comments ?

[19:09:43] <MrBigNum> in what situation you guys do not recommend use mongodb in a web application?

[19:16:49] <barque> are there any plans to grossly increase maximum BSON size in the future?

[19:17:15] <barque> also, how do I make a reference to another document from a document?

[20:36:06] <enigmeow> .../src/database/MongoDatabase.cpp:14:35: fatal error: mongo/client/dbclient.h: No such file or directory

[20:36:13] <enigmeow> is dbclient.h supposed to be part of the release