Log file Viewer
#mongodb logs for Thursday the 30th of April, 2015
[03:12:49] <arussel> GothAlice: thanks for your answer 'db.collection.update({arr: {$elemMatch: {a: "b"}}}, {$inc: {"arr.$.t": 1}})' is there a way to have it all in the update ?
[04:10:20] <mongodoh> if i wanted to replicate data from Oracle to MongoDB, what would be the best solution?
[04:12:24] <mongodoh> Oracle DB is for OLTP application but MongoDB would only need a subset of the data
[05:16:51] <mongodoh> joannac: found a decent json pl/sql library so it wouldn't be hard to create a document for the subset data from Oracle
[05:58:04] <wday> mms backup provisioning question - for existing large installations, do new shards need to be provisioned on the mms backup backend before being added to the cluster, or is it safe to add more shards and let mms backup pick up the change?
[06:18:53] <Boomtime> wday: when you select a cluster for backup in mms, it will resolve the shards, at all times
[06:19:19] <Boomtime> you _must_ not select individual shards for backup (like you would a replica-set) or this will actually block proper function
[07:50:48] <jeanre> is there a way to add two hours to $group: { _id: { hour: { $hour: "$stopTime" } }}
[07:52:11] <jeanre> because mongo uses UCT and we are in GMT my query is returning the incorrect date
[08:51:46] <newbsduser> how can I increase update operation count /per second... (current max update operation count 2000-3000 on single instance). I read about fast updates.. Are there another options to do that?
[09:05:53] <girb1> new to mongDB please help .. I'm dumping 30 days old data as JSON dump to file using pymongo
[09:08:00] <girb1> when I restore there is a different ObjectID in mongo .. is there a way I can restore the same ObjectID
[09:49:53] <pigoz> supposed I have a collection with an array field. I have a record with field: [a, b, c], I want queries with [a, b, c] or [a, b, c, d] to match, but for example: [a] should not match. Is there a way to model this?
[10:31:53] <KekSi> depends on what you want from it - i personally like robomongo but it's useless with my current setup (using ssl certificates to log in - which robomongo doesn't support yet)
[11:41:28] <bigon> we would like to upgrade from 2.0.2 to 2.6.x, can we just upgrade in one go? or should be upgrade with all the intemediate versions and restart in between
[13:46:17] <inad922> If I connect with client = MongoClient("mongodb://newrelic:kft7dx@127.0.0.1") it doesn't work even if I try to access the db via db["newrelic_test"] but if I connect via client = MongoClient("mongodb://newrelic:kft7dx@127.0.0.1/newrelic_test") it just works
[13:47:51] <ignasr> hello, I am trying to find why mongodb is saturating my disks with ~1000 IOPs every hour for one minute, when normally I have ~100 IOPs (DB size 6GB, RAM size 4 GB). How should I debug this? mongostat show locked > 100% and queue build up, but it seems that this is only the result of IO starvation.
[13:53:36] <ignasr> deathanchor: lots of slow queries logged for that minute, but again, its probably only the result of slow disk responce at that minute. There is no query count increase at that time
[14:22:35] <demahum> I am following this tutorial: http://www.learn2crack.com/2014/06/android-chat-application-using-gcm-server.html#, but I am stuck at the mongoDB part of it. I have created a database using robomongo, but now I should create a directory (you will recognize steps on the link). Any help?
[14:32:35] <StephenLynx> demahum you don't have to create databases or collections, you just use them and mongo will create them if they are not found
[14:37:56] <revoohc> Anyone know where to find the definition of what serverStatus.opcounters.command is representing? We just upgraded from 2.2 -> 2.6.9 and our command numbers have dropped significantly. All I’ve been able to find out is that 2.6 is no longer counter writes as commands. But what exactly is it counting as a command?
[14:44:16] <terlar> Anyone using the new Ruby Driver 2.0? I have been experiencing some weird issues.
[15:15:35] <eren> now secondary nodes are STARTUP2, and I see that one of them reports " clearing collection plan cache - 1000 write operations detected since last refresh."
[15:15:52] <eren> the other one doesn't report it but it reports: STORAGE [rsSync] clone ceilometer.meter 10647551
[15:40:26] <mmars> when I configure mongod v2.6 w/ replSet, what does that config option do? what if I run mongod with a replSet option but don't initiate a replication set or configure one?
[15:45:41] <gansbrest> hi. My logs seem to be filled with following messages http://pastebin.com/tDHaahcM
[15:56:23] <terlar> https://github.com/mongodb/mongo-ruby-driver/blob/master/lib/mongo/operation/executable.rb#L35
[16:34:15] <dorongutman> I have a collection of photos, in which every document has an object representing different sizes of photos: {_id: 11112222, sizes: { small: “http://blabla.com/11112222.small.jpg”, medium: “http://blabla.com/11112222.medium.jpg” …… etc..
[16:35:13] <dorongutman> is there an update query I can run that changes all collections with a “medium” size to a different url such as http://blabla.com/11112222.mid.jpg ?
[17:20:50] <Valesk> Did the C# Driver for Mongo remove FindAllAs()? Can' t seem to find any recent documentation on it..
[17:23:57] <Xeon06> Hey guys. Can anyone tell me if it would be a good idea to make sure that find queries like {foo: {$in: ["one item"]}} are changed to {foo: "one item"} or is that optimization already taken care of?
[20:11:31] <arussel> I have a list of object, {timestamp, value}, I need to push a new value and remove the value older than a specific timestamp
[20:58:01] <PVDM> Hey guys... I'm trying to see if I can do this with mongoose, but not getting much help in my google searches: I have object/collection A which references Object/collection StaticA, I want to limit my results in a query of A by something defined on StaticA...
[22:28:03] <PVDM> ahh, thanks. I came to the conclusion that we there was a better way to structure our data :)
[22:38:31] <GothAlice> PVDM: Due to the number of support requests I see involving Mongoose, it's my general suggestion to prototype data design and methodology using raw MongoDB tools instead of abstraction layers. This can make support easier, as well as eliminates distractions. MongoDB is schema-less, and making use of this for prototyping saves a lot of time.
[22:39:15] <GothAlice> (If you can't get it working in the MongoDB shell, it's doubtful adding more glue will improve things. ;)
[23:58:16] <Wulf> I've got a collection named `stats'. How can I see what's inside? db.stats.find() says "TypeError: Object function (scale){ return this.runCommand( { dbstats : 1 , scale : scale } ); } has no method 'find'"