Log file Viewer

[03:28:15] <someprimetime> if i’m connecting on one server through a mongodb driver (node.js based), do I need to have mongo installed on both servers or just the one that I’m connecting to?

[03:54:21] <phrozensilver> Can someone help me and tell me what I am doing wrong, is my data setup wrong? Trying to send an array of objects. The name field always gets sent - https://gist.github.com/rdallaire/241dd48262ec856620c1

[12:30:21] <Gargoyle> Do all members of a replicaSet have to be online when you run rs.init() rs.add(), etc?

[14:42:32] <jbelcher> Hi there. I'm seeking some outside input on creating new collection items on the client w/ optimistic updates. Right now we're basically creating a new object client-side, giving it a BSON ID we also generate clientside, and then POSTing that new object to our REST API.

[14:43:33] <jbelcher> Generating IDs client-side feels fragile. What are some other strategies we might employ? Some sort of a temporary token sent back in the response?

[14:44:48] <kali> jbelcher: how is that fragile ?

[14:48:54] <jbelcher> kali: I suppose it just seems natural to let the server handle UUID. Are there any real dangers to generating them clientside then?

[15:15:09] <kali> no

[15:16:03] <kali> jbelcher: http://stackoverflow.com/a/5694803/172537

[15:16:15] <kali> jbelcher: unless you really make it on purpose, it's safe.

[15:16:29] <kali> jbelcher: and i mean, you need to try hard

[15:22:14] <jbelcher> kali: ah, yeah that makes sense.Thanks for the link. Most client-side implementations of objectid I've seen fake the machine and pid with random and store those values in LS. I suppose that's more to the heart of it for me.

[16:43:21] <ernetas> Hey guys.

[16:43:28] <ernetas> I restored an instance from backups...

[16:43:42] <ernetas> I'm now trying to reinitiate a new replica set out of it, but it doesn't work...

[16:43:45] <ernetas> http://pastebin.com/v2kSGP7n

[16:44:10] <ernetas> What's the issue here? It still keeps old data about rs... How do I remove it from system.replset and so on?

[16:59:53] <kali> ernetas: stop the server, remove local* in the dbpath

[16:59:59] <kali> ernetas: and start again

[17:03:39] <phrozensilver> hey

[17:31:35] <ernetas> kali: last time I did this, it destroyed all my mongo databases... So, thanks, but no.

[17:37:50] <kali> ernetas: well, don't ask for advice, then

[17:39:01] <Ragnis> Hello. I'm trying to use the c driver, however simply including the mongoc.h header causes *a lot* of errors: http://pastebin.com/KsNbW69d I installed version 0.98.0 from GitHub.

[17:52:37] <Ragnis> Now I just feel stupid. I didn't properly set the include paths when compiling.

[18:39:16] <ernetas> kali: that's exactly why I'm asking.

[19:24:59] <Guest_> lol

[19:26:41] <desmondHume> just installed Textual as irc client on mac os

[19:30:45] <Guest_> .

[19:44:37] <bramgg> I'm having a very strange problem with Pymongo. When I upsert documents in my script it creates duplicates, but when I use the same upsert code in the Python console it upserts properly. Is there some sort of known issue with this?

[20:30:39] <progrock> I just install mongodb-org package on ubuntu 14.04. When I run sudo services mongod start I get: sh: 0: getcwd() failed: No such file or directory

[20:30:42] <progrock> start: Unknown job: mongod

[20:31:05] <progrock> can anyone point me in the right direction to get mongo working

[20:34:21] <ernetas> progrock: maybe "sudo service mongodb start"

[20:36:03] <progrock> ernetas: thats what I did

[20:37:35] <progrock> weird now I'm getting a different error, that makes no sense, lol.. now its jsut unknown job: mongod... not sure how that changed

[20:41:18] <ernetas> progrock: check init scripts, see what it tries to run and so on.

[20:44:49] <progrock> yeah that's what I'm doing now, and making sur ei have the right user and group on folders

[23:20:53] <huleo> hello

[23:22:12] <huleo> wondering at how much hassle I'm looking at, maybe, just /maybe/ there is a magic pill...so we have posts, in "posts" collection, and users, in "users" collection...each "post" has "author" field referencing user that created it

[23:22:51] <huleo> now, what seems to be the simplest way to get, let's say, 10 most active users - the ones that wrote most posts

[23:23:39] <huleo> actually, forget about limiting results to "10" users, that's the simplest part