PMXBOT Log file Viewer

Help | Karma | Search:

#mongodb logs for Friday the 3rd of January, 2014

(Back to #mongodb overview) (Back to channel listing) (Animate logs)
[00:17:18] <unholycrab> joannac:
[00:17:28] <unholycrab> { "errmsg" : "version number must increase, old: 95594 new: 11", "ok" : 0 }
[00:17:51] <unholycrab> if i do rs.reconfig(cfg, {force: true})
[00:17:56] <unholycrab> i get a random version number
[00:18:04] <unholycrab> "errmsg" : "version number must increase, old: 95594 new: 91406",
[00:23:34] <unholycrab> document version 95594 was never the active config, and i just deleted it
[00:23:45] <unholycrab> how do i make it forget about 95594
[01:16:24] <love-cash> ok
[01:16:43] <love-cash> when copying files to node 2/3 and 3/3 from 1/3,
[01:16:54] <love-cash> do I delete local* on 2/3 and 3/3
[01:16:54] <love-cash> ?
[01:17:07] <love-cash> before rs.add() 2/3 and 3/3?
[01:17:18] <love-cash> http://docs.mongodb.org/manual/tutorial/restore-replica-set-from-backup/
[01:17:20] <love-cash> is unclear
[01:17:24] <love-cash> on this point
[01:25:01] <unholycrab> forced current document to version 95594 and worked around the problem
[01:42:06] <love-cash> are mongo files best rsynced with -z option?
[01:42:12] <love-cash> are the pretty compressible?
[01:58:22] <love-cash> few answers here
[01:58:23] <love-cash> bad
[01:58:27] <love-cash> docs bad
[01:58:37] <love-cash> not sure about the choice to use this project
[01:59:22] <cheeser> try asking during the EST day
[06:07:24] <zhodge> if I want to overwrite an entire document (given an _id), collection.save is the appropriate call, correct?
[06:07:35] <zhodge> (I'm using the official node.js driver if that makes a difference)
[06:10:11] <zhodge> this is what my call looks like along with resultant output https://gist.github.com/qcom/8233617
[06:10:25] <zhodge> the id I pass into the query document object literal is just a string
[06:10:32] <zhodge> i.e. not an ObjectId("withMyStringHere")
[06:10:34] <zhodge> does that matter?
[06:34:04] <zhodge> ah
[06:34:05] <zhodge> that does matter
[06:34:16] <zhodge> var ObjectID = require('mongodb').ObjectID
[06:34:38] <zhodge> collection.save({ _id : new ObjectID(myStringIDVariable) })
[07:53:41] <Happy_Meal> Hello guys, I'm back again. I have a problem. I want to do something like Craigslist (except I've made it easier to use). I was told not to use skip and limit for counting pages, but instead to utilize the ID field (since it's indexed). My problem is, there's no way to keep postings from running into each other. If I serialize the _id field, I will have to make a count connection to the server, then increment the count by one. By t
[07:53:41] <Happy_Meal> his time, someone else could have already posted to that number slot.
[08:18:03] <jkitchen> Happy_Meal: no, just let mongo add its own _id, and then query for things >= id limit blah
[08:18:11] <jkitchen> rather, > id
[08:25:51] <Happy_Meal> OK thanks
[14:44:10] <tiller> Hi!
[14:45:16] <tiller> I've an error that I don't understand with my Java program :
[14:45:18] <tiller> com.mongodb.MongoException: have to have sort key in projection and removing it
[14:45:50] <tiller> with: http://pastebin.com/4etzYhun
[14:45:58] <tiller> But I've never had any problem with this before
[14:46:15] <tiller> The only thing I changed is that I sharded & replicated my db (I think)
[14:56:33] <tiller> And I don't understand, because the same query work in the console..
[14:56:35] <tiller> mongos> db.sslInspections.findOne({serviceId: ObjectId('52af46e9e4b0599a6398b716')}, {firstDate: 0, lastDate:0},{lastDate:-1});
[14:56:39] <tiller> > null
[14:57:07] <the_lord> Hello, I'm trying to make a unique index for embedded documents
[14:57:29] <the_lord> and it's not working at all, I don't know what I'm doing wrong
[15:44:48] <sweb1> i want to know what's disadvantage of mongodb. i write an web crawler like google reader for wrapping some news websites. i'm using mysql but i have some problems like finding related news with matching tags (tables : news, tags, news_tag)
[15:45:19] <sweb1> i need write performance also data durability
[15:45:35] <sweb1> i wanna know is mongodb good choise for me or not
[15:46:08] <sweb1> also i have websocket layer for live data intraction ... also this is my another situation
[15:46:15] <sweb1> is mongodb good for my usecase or not
[15:48:53] <kali> i'd say... not significantly better or worse than mysql
[15:50:37] <sweb1> kali: i'm using percona instead. but in rdbm it's good for me
[15:51:07] <sweb1> kali: what do you think about my usecase ... is there any wiki for disadvatage of mongodb ... i wanna see theire usecases
[15:52:19] <ppetermann> isn't there a comparison on the mongodb website?
[15:55:45] <sweb1> ppetermann: i cant see any useful data
[15:56:19] <ppetermann> sweb1: since i have no idea what "any useful data" means, i guess i can't help you then
[15:57:28] <sweb1> ppetermann: i cant find anything in mogodbwebsite that answer the question ? what's mongodb good for ? and more important. what's mongodb bad for ?
[15:58:03] <ppetermann> http://www.mongodb.com/learn/nosql
[15:58:06] <ppetermann> http://www.mongodb.com/use-cases
[15:58:32] <kali> sweb1: mongodb is a general purpose database. it can pretty much do anything a rdbms can do (hell, turing completeness and all). it's better at scaling, less good at transaction.
[15:58:35] <ppetermann> and basically: its good for everything not relational, and bad for everything where you need relational data
[15:59:01] <ppetermann> kali: i hate this "better at scaling" claim - it scales differently
[15:59:30] <ppetermann> and "less good at transaction" bascally means, if you don't use tokumx (a fork of mongo) you don't get transactional safety
[15:59:48] <kali> ppetermann: and i hate the "bad for everything where you need relational data" :P
[16:00:00] <kali> ppetermann: the question is moronic. what do you expect for an anwser ? :)
[16:00:46] <ppetermann> i have not asked any question
[16:01:04] <kali> sweb1: http://stackoverflow.com/questions/1476295/when-to-use-mongodb-or-other-document-oriented-database-systems the first anwser is quite good.
[16:01:29] <kali> even if it's a tad old
[16:01:37] <ppetermann> and basically there is no such thing as a "moronic question", there might simply be a lesser-good-presentation of the question.
[16:02:18] <kali> ppetermann: ok. i'll rephrase that. it was not a question for irc, it was a question for goodle.
[16:02:22] <kali> google
[16:04:00] <ppetermann> i think the first link i answered with is actually quite good in answering what he wants to know, but its stupid to find through google
[21:21:31] <astronouth7303> are there any packages or frameworks that allow for a live feed of updates to MongoDB?
[21:22:05] <astronouth7303> meaning that my application gets notified when something changes in Mongo