Log file Viewer
#mongodb logs for Wednesday the 28th of October, 2015
[00:09:42] <StephenLynx> https://gitlab.com/mrseth/LynxChan/blob/master/src/be/engine/gridFsHandler.js here
[10:14:17] <LukeOne> When importing big CSV files with 500k rows each, mongoimport confirms the import and the correct count but if I'm counting the collection around 750k entries the collection doesn't change the count anymore.
[11:16:46] <kephu> I'm trying to figure out a thing with aggregation: I've got a bunch of documents with performance variables (say, user,sys,idle). I'd like to be able to get an array containing all the data from those variables, as arrays themselves
[11:25:41] <acidjazz> hey everyone, can i get a shotout of everyoens favorite mongodb GUI? so far mongochef has been my favorite
[11:47:10] <kephu> ok so far, my best guess for my problem was to make an array-like object, with attributes like 'data.0'
[11:51:47] <aps> If I run db.fsyncLock() more than once, I'm unable to db.fsyncUnlock(). Is this a known issue? How do I Unlock now? :/
[11:57:24] <acidjazz> StephenLynx: i mainly like it becuase w/ guis im jusyt verifying and moving data
[11:59:08] <StephenLynx> cheeser this guy is advertising something, is that allowed by channel rules?
[12:09:53] <acidjazz> StephenLynx: i was to see if you were trolling, also if you use the terminal alot you should check out mongo-hacker
[12:11:05] <StephenLynx> >This will rename your existing .mongorc.js file to .mongorc.js.orig and link mongo_hacker.js to .mongorc.js in your home directory
[12:12:42] <StephenLynx> MongoDB Database Server and Tools Free Software Foundation’s GNU AGPL v3.0.
[12:14:06] <StephenLynx> and excuse me for having standards for software, it doesn't mean I am insecure, just that I don't run any random crap I find on the internet.
[13:39:19] <christo_m> hello, im trying to query a subcollection.. namely, an item that has a user in it .. but when i try to do a query like { 'user._id': objectIDHERE } it fails to return
[13:39:44] <christo_m> how should i do the query? ive tried nesting it also like { user: { _id: objectIDHERE } }
[14:07:22] <amitprakash> Specifically, I want to log all mongodb queries without impacting execution times.. How can I achieve this?
[14:40:07] <yopp> Hum. I've got a collection with two indexes: on _id and on another field (something_id). It got 4 records. And I'm seeing this: http://pastie.org/private/4oldt2r3umt0zj7bvduta
[15:32:02] <christo_m> ive read the docs , tried this: http://stackoverflow.com/questions/28665271/find-subdocument-by-id-inside-a-array-field-of-nested-object , still won't come up in results
[15:32:03] <obiwahn> Is there #mongodb-users or something? I see people joining and asking something but not many get a response
[15:38:13] <christo_m> hello, im trying to query a subcollection.. namely, an item that has a user in it .. but when i try to do a query like { 'user._id': objectIDHERE } it fails to return
[15:38:18] <christo_m> how should i do the query? ive tried nesting it also like { user: { _id: objectIDHERE } }
[15:38:24] <symbol> obiwahn: There's a users group on google that's helpful, http://groups.google.com/group/mongodb-user/
[15:44:17] <symbol> It appears that Mongoose should be handling that for you. christo_m: run the same query in the shell and see if it returns any results.
[15:44:35] <symbol> christo_m: http://stackoverflow.com/questions/7878557/cant-find-documents-searching-by-objectid-using-mongoose
[15:46:24] <deathanchor> I have field called total, for the $group how do I aggregate the $sum of total into a single result?
[15:47:28] <christo_m> trying: db.queueitems.find({'user._id': '5630ebe9d47e87d777c9c22e'}, {'user.$': true})
[16:01:39] <pokEarl> Hey Friends. so I am partially responsible for a mongoDB and am about to update a form of Schema so you can look up to see all the different fields and documents (And I guess maybe have some information about fields that might still be present but are deprecated/being phased out f.ex?) any suggestions on best way to make this so its both easy to edit/update, and easy to navigate/take in visually?
[16:22:53] <fakingfantastic_> How do you do an order_by on a field that may not be on every document?
[18:55:09] <shlant> hi all. How should I handle permissions for mongo ssl certs? I was putting them in /etc/ssl/private, but mongo doesn't have access. Should I put the combined pem file in /etc/ssl and only give mongo read permissions?
[19:25:29] <christo_m> db.queueitems.find({'user._id': ObjectId('56311a292625f11f80e6cf4b'), 'toWatch': true}) this query seems to work in mongo console
[19:45:47] <shlant> is there something different I need to do when adding a replica set member when using TLS? or still rs.add("blah")?
[19:46:28] <shlant> but I can hit the hostname with dns. Does rs.add have TLS specific errors? or just the generic "connection attempt failed"
[19:51:16] <mylord> how/can I update only a subdocument value, and not run over the entire document, when doing update?
[19:52:38] <GothAlice> {$set: {'field': value}}, or, with embedded documents, {$set: {'subdocument.field': value}}, etc.
[20:06:49] <mylord> what are the memory and other resources costs of using a separate collection, by default?
[20:07:37] <mylord> 1 per each airline company in the world.. 1000s.. not millions like clients/users
[20:10:20] <galeto> What minimum set of indexes would ensure this query use a good plan ? (mongo 3.0) https://gist.github.com/AlexGalays/e3bd4ee9e1845fdcb062
[20:10:54] <jrdn_> is this the most optimal way of updating / inserting into an array of embedded documents? https://gist.github.com/j/1f2a4b6942d5abb16971
[20:11:11] <jrdn_> it feels weird doing that, but i don't want to create another collection for the sake of upserting... or should i?
[20:13:29] <galeto> jrdn_ : Maybe you don't need it, but that way of doing upserts is of course not fully safe. findAndModify is your best bet
[20:20:31] <GothAlice> findAndModify isn't safe either, really. (Unless you memorize the documentation page for it and all of its many side-effects, and code really, really carefully.)
[20:21:38] <GothAlice> jrdn_: Determining if it's optimal or not would require measurement/benchmarking, but the pattern you're using is one I frequently use, too.
[20:22:17] <GothAlice> Ref: the forums I wrote that store replies to threads that way, and these helper methods I wrote to manage the nested data: https://github.com/bravecollective/forums/blob/develop/brave/forums/component/thread/model.py#L99-L161
[20:23:13] <GothAlice> jrdn_: findAndModify is a particularly strange case. Despite needing the basic pattern of "get something, update it" and/or "update something, then get it", I've never once been forced to use findAndModify.
[20:24:39] <jrdn_> yeah, i want to say that I haven't used it either. in my case, it's (and i'm assuming right now since we're not live) not something that will be getting executed a lot.
[20:26:30] <deathanchor> if you need the info. find(), if you need to update it, update(). why would you need both is beyond me.
[20:27:12] <jrdn_> well when using safe updates, doesn't it return whether the document was modified or not, and if it wasn't modified, doesn't it also say that the document was found, but not modified
[20:29:15] <GothAlice> On the gripping hand, the vast majority of my upserts are pre-aggregation fire and forget queries.
[20:30:51] <jrdn_> i've never really used a ton of arrays of embedded documents, so i was lost at the minimum and most optimum efforts on adding a document to an array and making sure it's unique
[20:33:46] <deathanchor> it's fine if you have a finite model, but most people just keep piling things into a array or sub doc and then hit the doc size limit.
[20:36:50] <jrdn_> last time i didn't do it and put it into a different collection (way easier to just move forward, btw), then people start talking about how you're an idiot for not using SQL.
[20:42:35] <GothAlice> jrdn_: Eh, yeah, but only because I'm dropping out of MongoEngine, then shoving the results back in. ;P
[20:43:12] <GothAlice> https://github.com/marrow/task/blob/develop/marrow/task/queryset.py?ts=4#L28-L81 is a much, much prettier example of ME customization. ;)
[20:43:43] <jrdn_> yeah i had to go to the documentation to see that the operations are followed by the underscore and field
[20:44:07] <GothAlice> Indeed; I didn't duplicate that documentation. It's "Django-style". field[__subfield[…]][__operation]=value
[22:13:47] <richard_w> my documents have the structure {_id: "...", name: "...", values: { "key1": "val1", ... }}
[22:14:39] <richard_w> when i try to replace them with new documents the children of "values" get replicated
[22:15:50] <richard_w> i am using reactivemongo in scala: "collection.update(BSONDocument("name" -> princ.name), princ)"
[22:17:31] <richard_w> using "BSONDocument("_id" -> BSONObjectID(princ.id), "name" -> princ.name, "values" -> princ.values)"
[22:21:00] <StephenLynx> no idea about any of that. if you give me a query valid on the shell, I might be able to help
[23:25:49] <tejasmanohar> And with the multi parameter set to true with the findAndModify query, this is identical to update() right?
[23:27:56] <BurtyB> tejasmanohar, I wouldn't call them the same.. findandmodify can be used for different things than update
[23:28:59] <tejasmanohar> also, anyone familiar w/ https://github.com/Automattic/monk ? and do you know if it exposes document#functions like save() or does it only expose collection#functions like update, find, whatnot