Log file Viewer
#mongodb logs for Friday the 17th of July, 2015
[00:41:44] <johnisbest> Hey, if I use one giant mongo document, is there an easy way to only return a value two arrays deep? For example: http://pastebin.com/3KugYku9 Can I use a mongo query to find a pet object by the pet's ID?
[00:43:21] <johnisbest> I thought I could use the aggregation pipeline and use a redact but I do not know if that would work here
[00:44:10] <johnisbest> Also I want to only return the pet object and not most of the other data. I am hoping I dont need to waste time with looping through arrays to find the object
[03:13:45] <aldwinaldwin> goodmorning, goodevening, ... I have a question about a shard key that I try to use
[03:14:12] <aldwinaldwin> more specificly, what is not really adviced, but just trying to use an ISODate as shard key
[03:14:26] <aldwinaldwin> the documents are already created, the index {datetime:1} is created, all documents have a value for 'datetime'
[03:14:40] <aldwinaldwin> sh.shardCollection("testing.test",{datetime:1}) => "errmsg" : "found missing value in key { : null }
[03:14:59] <aldwinaldwin> { "_id" : ObjectId("55a86b5e8d79543ae1bf5989"), "info" : "blablabla", "unixtime" : "2015-07-17 02:41:34.797672", "datetime " : ISODate("2015-07-17T02:41:34.797Z") }
[03:16:13] <aldwinaldwin> so... I don't really understand why the shardCollection function says there is a null value in the datetime
[04:01:43] <aldwinaldwin> joannac: thank you soooo much, pfff, hours lost on this. oh well, lesson learned
[06:51:21] <johnflux_> > db.actionlaunch.find({"payload.rank" : {$exists : true}}).forEach( function(obj) { obj.payload.rank = new NumberInt( obj.payload.rank ); db.actionlaunch.save(obj); } );
[08:24:14] <vagelis> Hello, I have 2 fields for Date. Created and modified. If 1 document has created doesnt have modified and if 1 document has modified doesnt have created. How can i sorted these documents by Date ? I just use multiple keys in the sort? modified: 1, created: 1 ?
[10:10:13] <penthief> Hi, I have a CodecConfigurationException pasted here: http://pastie.org/10298155 Could someone tell me what I am doing wrong?
[10:52:56] <m4k> I have this http://pastebin.com/548CiKEj document How do I filter by property_type I tried this db.meta.find({"specifications.$.property_type":"commercial"})
[11:37:54] <m4k> I have this http://pastebin.com/548CiKEj document How do I filter by property_type I tried this db.meta.find({"specifications.$.property_type":"commercial"})
[11:54:14] <Soapie> Hi. I'm fairly new to MongoDB (I'm from a SQL background). I'm using the c# .net driver and having trouble working out some of the queries. Am I in the right place to get help with the C# .NET MongoDB driver?
[11:58:23] <cheeser> kind of. not many of us are c# users from what I can tell. i work with the c# driver guys so I have a passing familiarity so might be able to help. the users list is probably your best bet but we'll do what we can here.
[12:01:03] <m4k> jelle: There is one document specifications and it has many subdocuments as dict. So Its not working with $in
[12:01:55] <dijack> one question...what's the difference between Priority 0 member not able to TRIGGER an election and being ABLE TO VOTE in an election?
[12:02:57] <m4k> jelle: I tried this db.meta.find({"specifications.property_type": { $in: ["residential"]}}) and db.meta.find({"specifications.$.property_type": { $in: ["residential"]}})
[12:12:00] <Soapie> Thanks @cheeser. At the moment, I'm struggling to do what, in sql terms, would be a datediff, within an aggreagtion. Basically I'm trying to get the number of days elapsed since the last event that a user attended. So you'd see a list of user records with the number of days since they last attended an event.
[14:19:39] <amitprakash> Hi, http://docs.mongodb.org/master//release-notes/2.6-compatibility/#index-key-length-incompatibility tells me that repairDatabase will fail on encountering the index too large issues
[16:58:20] <zwarag> Very noob question here. Filling a collection with the value of my cpu temperature every second. How do I to keep an average of the last 60 entrys stored?
[17:41:07] <daidoji1> zwarag: however 60 items is small, just make a capped collection, select all the data and average in a script?
[17:42:35] <daidoji1> can anyone help me improve this query? It looks like its taking a while to create my export file
[17:42:37] <daidoji1> nohup mongoexport --host mongotest.data.revup.com -d revup -c st --query '{date: {$gte: Date(1420070400000)}}' 2> st_export.log | xz > /mnt/st_export_$(date "+%Y%m%d").xz &
[17:50:15] <amitprakash> Hi, http://docs.mongodb.org/master//release-notes/2.6-compatibility/#index-key-length-incompatibility tells me that repairDatabase will fail on encountering the index too large issues
[17:50:39] <amitprakash> How can I repair an old database that throws up too large to index errors on repairDatabase?
[18:00:22] <diegoaguilar> Hello Im trying mongo --quiet production-season-lf-199-2015 --eval 'printjson(db.groupg18.find({_id: ObjectId("55a550103711fed46621fd97")}, {calendar: true}).pretty())' > c.js
[18:19:08] <daidoji1> try db.groupg18.find({_id: ObjectId("55a550103711fed46621fd97")}, {calendar: true}).forEach(printjson)
[18:19:36] <daidoji1> the way you're doing it is calling printjson on the cursor, which just prints out the cursor definition probably
[18:33:18] <daidoji1> nohup mongoexport --host mongotest.data.revup.com -d revup -c st --query '{date: {$gte: new Date(1420070400000)}}' 2> st_export.log | xz > /mnt/st_export_$(date "+%Y%m%d").xz &
[18:33:41] <daidoji1> this fails to return all records (only returning about 20k). Would someone mind helping me with what I may be doing incorrectly?
[18:34:51] <blizzow> I upgraded my sharded cluster to 3.0.4 yesterday. After the upgrade, the load average on my secondaries for one shard is regularly going ballistic. I fired up mongotop and see a few collections returning really high read times (30000ms-110000ms). I see some query against one db.collection that's taking a long time...is there a way to figure out what the source of that query is?
[18:34:55] <blizzow> I have 4 CPUs on my mongod instances and both secondaries are pegged at 400%CPU.
[18:35:02] <daidoji1> thats weird, when I run it in console with explain I get this error https://gist.github.com/daidoji/59c5fd1b3a3cac4f2fc6
[18:39:16] <daidoji1> http://stackoverflow.com/questions/7958228/mongodb-running-but-cant-connect-using-shell
[18:45:13] <deathanchor> the server is backward compatiable to older clients, but clients aren't backwards compatible to older servers
[18:46:26] <daidoji1> deathanchor: okay I'll take your word on it but it seems weird they would make that decision
[18:48:42] <daidoji1> and thats why its weird. There's nothing I want to do more with an enterprise grade database server full of my business's most important information than upgrade it all the time.
[18:49:04] <daidoji1> whereas clients are lightweight and can be replaced easily in the typical scenario
[18:51:48] <daidoji1> deathanchor: anyways thanks for your help, that probably is the issue come to think of it
[18:52:01] <deathanchor> I'm not sure, but if I have issues, I usually verify that I'm using like for like versions server/client.
[18:59:45] <deathanchor> for example you are _id : ObjectId("5594c7141fde007c80c6b84f") in our people.ToBrainWash db/collection.
[21:30:20] <rusty78> If I want to find the top 10 highest numbers in a specific field in a collection of documents is there a built-in query for this?
[21:30:40] <rusty78> like I have 100 documents in a collection all with the field prices, and I want to get the 10 highest prices
[22:16:41] <rusty78> is it a bad idea to index a collection of sessions, since documents get added and removed so quickly?
[22:29:54] <rusty78> it's actually to store socket connections - basically when a user enters the website they connect to a socket
[22:30:09] <rusty78> I store their socket.id and username as a document on the collection,a nd when they leave I delete it
[22:30:37] <rusty78> but at the same time they get created/moved very fast, so would indexing cause a large performance drop if I used one?
[22:46:59] <falieson> I want to build my own analytics, reporting, and logging features for my app but I'm trying to decide if I should keep it in the same db? AnalyticsPages AnalyticsUsers AnalyticsEvents AnalyticsLogs or should I just have 1 collection Analytics? and do something like record_type: "pages, users, logs, events,"
[22:48:41] <falieson> i realize there's multiple questions about should I have the app and the analytics share the db, and if i should split parts of the analytics module into their own collections
[23:44:10] <falieson> daidojil it being analytics type data is mostly write once and view aggregates