Log file Viewer
#mongodb logs for Monday the 27th of April, 2015
[06:49:57] <thomaschaaf> hello I just upgraded from wheezy to jessie today on one of our servers. Now I cant start mongodb anymore: http://pastie.org/10115694
[08:12:33] <errakeshpd> how to use rails and mongo ( Model.find(1) is for rails and Table.find(1) for mongo ) so it may make some confusion while seeing the code
[08:21:46] <eren> hey folks, I switched to mongod 3.0.2 but I forgot to switch wiredTiger storage, it is still using the old backend
[08:28:08] <eren> Derick: replica set could be a good solution. Unfortunately I have 1 mongo installation and it's actively used now, collecting meters from agents. I believe the collector cannot process them and write to mongo fast enough and the messages are queued in rabbit
[08:37:43] <eren> Derick: yeah, I have 2million messages in rabbit, that's way more than it shuold be :)
[08:38:36] <eren> I believe I will go with replica solution. I will spin up a new replica with wiredTiger, add this replica to single mongo instance, wait for sync, and ... ?
[08:39:23] <eren> should I switch to this replica as a main mongo database, or wait for the sync. I have 60GB of data currently
[11:00:36] <amitprakash> Hi, for queries in currentOp which aren't being recorded by the virtue of being too large, is there a way for me to inspect what this query is?
[11:01:22] <eren> how can I instruct mongodb to rebuild indexes? I moved to wiredtiger via "mongodump/mongorestore" and I see that /var/lib/mongo is 350MB, where in the previous instance it was 7GB
[13:24:21] <mtree> is it good practice to disable authentication requirement when automatically testing protected endpoints?
[13:41:29] <snowcode> I've executed a $geoNear query. Everything it's okay until I add as query param the sort key:
[13:41:53] <snowcode> query execution fails with this error: Can't canonicalize query: BadValue unknown top level operator: $sort
[14:03:38] <snowcode> StephenLynx Thank you :) Have you ever worked with $geoNear? I've tried to perform a query to get all the points near a [lon,lat] point but it returns wrong results (in fact it seems to return any results): https://gist.github.com/malcommac/72a689e6b8847fc5e289
[15:40:18] <MadLamb> can i could the number of properties of a object in mongo? {1: "something", 2: "something2"}
[17:15:24] <MadLamb> the docs say it should return a WriteResult object, but i'm receiving undefined
[17:21:12] <MadLamb> none, it returns as the comand suceeeded. Then i tried to get the WriteResult as debug, but its returning undefined instead.
[17:26:37] <MadLamb> StephenLynx, i'm running that directly in terminal as one-line-command. I just formatted it for readability purposes to show to you.
[17:27:32] <StephenLynx> ok, so you open your terminal, you type "mongo", then you type all that code?
[17:31:43] <MadLamb> the weird part is that it still returned undefined, but the command worked. Maybe the WriteResult is returned in terminal instead of the default method return?
[17:34:32] <snowcode> is possible to have an aggregate query with multiple params? I've added a $geoNear but I would to add some other conditions
[17:34:55] <StephenLynx> aggregate are designed for that purpose, you add operators in any amount or order.
[17:35:44] <snowcode> StephenLynx I've added another params to db.earthquakes.aggregate({"sourceID" : "444", $geoNear : {near : {type : "Point", coordinates : [13.369729,42.324681]},maxDistance : 20000,spherical : true,distanceField : 'distance'}})
[17:35:59] <snowcode> it returns error: ""exception: A pipeline stage specification object must contain exactly one field."
[17:58:29] <Takumo> Somewhat dumb question, what's best way to do a Mongo version of an SQL group and count?
[18:00:39] <StephenLynx> but if you need to perform complex operations to get the desired count, then you will need to use aggregate to count.
[18:03:59] <snowcode> StephenLynx and $match can contains $lte,$gte keywords? {$match : {time : {$gte : "2015-03-23T00:00:00Z",$lte :"2015-03-23T23:59:00Z" }}}
[18:19:53] <Takumo> StephenLynx: awesome! Let's see how it works on this dataset I'm adding to at about 100 docs/s
[18:33:32] <ToeSnacks> I have been draining a shard in my cluster for 3 days and the remaining chunks count has not gone down at all, how can I verify Mongo is actually draining and not in a hung state? Watching the logs on the primary of the draining shard shows it migrating data, but it's taking way too long.
[18:46:22] <StaticIP> Question: How would I connect 2 collections into 1 and have access to the information for use ona view?
[19:26:08] <dj3000> hi. I get this error when trying to install mongodb on ubuntu: W: Failed to fetch http://repo.mongodb.org/apt/ubuntu/dists/vivid/mongodb-org/3.0/multiverse/binary-i386/Packages 404 Not Found
[19:32:47] <StaticIP> daidoji, sorry for not clarifying. I want bring all the data in 2 collections. So say I have 1 collection called Users that has information about the user. 2 collection is called Events and in here it also has data. I want to connect to both and collect all the data and then in my views I want to display it.
[19:35:10] <StaticIP> it would be equal to say sql where you do a JOIN and collect information from 2 tables.
[19:37:09] <StaticIP> ok.. can i have a reasoning? or perhaps a work around that would be doable. because say, I want the user to see all the events he/she has been to.
[19:41:57] <StaticIP> ok how about this, how would i reference 2 collections to a user? or would the best way to do this would to have a collection called user and then inside of there have Events with an array of the events?
[19:42:50] <ToeSnacks> cheeser: do you know how to find out what/if mongo is moving from one shard to the other?
[19:50:51] <StephenLynx> StaticIP in that case you would need to have dynamic creation of collections.
[19:54:08] <StaticIP> i was thinking that i am setting this up incorrectly to mongodb. i come from a php/mysql background and thought i could use some of the building process like i would there into here. but i guess i have to figure out a different way of doing it.
[19:56:10] <StaticIP> i think, if i just use "Users" for all the storing data, meaning if there is an event that a user will be at, that information would go into Users document as Events (array)
[19:58:31] <StaticIP> and then when I want to display in a view for the Events for that particular User, it would just be all there. Am I thinking that correctly? Lol
[19:59:23] <StaticIP> I just started with node/express/mongodb, seriously like last week. I have gotten far in my web app but there were certain aspects I wasn't thinking because of my thinking in sql.
[20:01:29] <StephenLynx> and if you have a REALLY big amount of data, each document can only holds 16mb. which is millions and millions of words, but its a hard cap.
[20:04:21] <StephenLynx> for generic business systems I don't believe mongodb to be the best option, TBH.
[20:05:00] <StephenLynx> first theres just too much relations, then you have how often business rules change
[20:08:11] <StephenLynx> On the other hand, I would still use io.js, been using it with mysql and its pretty great.
[20:21:21] <ToeSnacks> are there any causes for 'ns not found, should be impossible' other than config desync?