Log file Viewer

[02:02:03] <patrickod> is it possible to use stored procedures in the aggregation framework?

[02:03:11] <patrickod> I'm looking to get the standard deviation across a field in the document set but it seems that isn't built-in command in the frameork

[02:03:25] <patrickod> and I'd rather not have to transfer the data out of mongodb to calculate it client side

[03:01:44] <lazyPower> i have a document in mongo, and the script that populated the entries perserved spaces in the key names. I need to run an update to replace them with an easier key to map. db.joblisting.update({ "Years of Experience": {exists: true} }, { $rename : { "Years of Experience" : "experience"} })

[03:02:09] <lazyPower> what is wrong with my query? It runs yet the key is never updated. have i missed something obvious?

[03:09:45] <crudson> exists -> $exists

[03:10:37] <crudson> and you want ({}, {} false, true) to update multiple documents

[03:11:44] <lazyPower> Doesn't appear to be working :|

[03:12:14] <crudson> create a paste of a source document so we can look at what is happening

[03:13:17] <lazyPower> https://gist.github.com/3764985

[03:14:22] <lazyPower> wait that worked

[03:14:29] <lazyPower> i missed a comma after the second object set

[03:19:18] <lazyPower> Thank you crudson

[03:19:48] <crudson> great :)

[03:41:43] <mindlogger> guys, does mongodb scale as good as Disco?

[03:41:52] <mindlogger> and is it faster?

[04:53:05] <jiffe1> do I need to do something special to make a replica set work when one half is down?

[12:13:03] <federico3> Anybody here using the Debian Pymongo packages? If so, have you ever impacted bugs.debian.org/cgi-bin/bugreport.cgi?bug=688040 ? Thanks

[12:18:21] <algernon> federico3: works fine here (sid, amd64)

[12:18:52] <algernon> oh. it does dump core, at exit time

[12:19:20] <federico3> same here, on one amd64 system out of two

[12:19:36] <federico3> it segfaults when the Python interpreter exits

[12:22:25] <federico3> thanks algernon

[12:27:26] <algernon> hmmm

[12:27:43] <algernon> recompiling with debug symbols for python2.6 only gets rid of the sigsegv

[13:20:47] <algernon> right, it is py2.7 specific, by the looks of it.

[13:26:26] <federico3> algernon, and on some amd64 hosts it does not happen :-/

[13:27:34] <algernon> it happens reliably here, got a backtrace too, though not too useful.

[13:32:47] <federico3> actually I'm having segfault with both 2.6 and 2.7

[13:36:22] <algernon> yeah, got a 2.6 segfault too, but that one I found less reliably reproducable

[13:37:09] <algernon> interesting though, if I compiled without python2.7-dev installed, I was not able to reproduce it at all

[13:37:17] <algernon> neither with 2.6, nor 2.7

[13:37:36] <federico3> here the segfault always happend with .6 and .7 on one host

[13:37:38] <federico3> and never on another

[14:01:59] <federico3> algernon, segfaulting in visit_decref()?

[14:02:06] <algernon> yes

[15:01:56] <federico3> algernon, did you got a segfault during the import?

[15:02:21] <algernon> federico3: nope, at exit

[15:02:55] <federico3> thanks

[18:55:48] <eldar> hi all! I'm new to mongo and trying to use TTL, but it doesn't quite work for me http://pastebin.com/hg145zL2

[18:56:05] <eldar> the documents still remain in the collection after 120 seconds

[18:56:13] <eldar> is it something wrong that I'm doing?

[22:08:26] <nemothekid> So, I am having issue where our mongos & mongos instance on the same machine will just both exit at the same time. The log files show that both instances quit when they receive a failed moveChunk because of "Invalid BSONObj size: -286331154 (0xEEEEEEEE)." However this is after days of receiving the issue (we are aware of it, and the only fix is to restore 2 near 1TB size machines which has been going on for the past couple days as

[22:08:26] <nemothekid> indexes are rebuilt). I'm not concerned about the move chunk error, but is it possible mongo has a max failures options that is killing the processes?

[22:33:30] <noordung> Hello guys... Latest Mongoose supports done in middleware, right?

[22:35:05] <noordung> Huh, I seem to be doing something wrong by calling next()... anyhow, NVM.