pmxbot IRC Log Viewer

[14:41:51] <hsk3> When I run "mongo" on my Linux (with "mongod" running in the background of course) and create a database, add collections, documents, etc.

[14:42:07] <hsk3> Can any other user on that Linux system also access that database I created by just running "mongo"?

[14:42:13] <hsk3> Or is what I create tied to my Linux user?

[14:45:50] <teprrr> you can enable remote access somewhere I think, per default it isn't listening remote connections (I hope)

[14:46:17] <hsk3> teprrr: i mean different local users. not remote connections

[14:46:19] <teprrr> then you do just mongo 123.123.123.123

[14:46:36] <teprrr> oh. I thought there's no differentiation between users on that level

[14:47:21] <hsk3> so if i create an ssh account on my server for my coworker, he automatically gets access to my mongodb databases?

[14:47:27] <hsk3> strange..

[14:51:11] <teprrr> per default so, I think

[15:02:50] <StephenLynx> hsk3, that is controlled by file permissions.

[15:03:12] <StephenLynx> so for the sake of convenience, mongo implements database users.

[15:03:28] <StephenLynx> so you can specify which db users can do what.

[15:03:53] <hsk3> StephenLynx: to do this i gotta start mongod with --auth, right?

[15:04:24] <StephenLynx> I think so.

[15:04:28] <StephenLynx> however

[15:04:43] <StephenLynx> I recommend setting that on the settings and running mongo though its service.

[15:07:17] <hsk3> StephenLynx: which settings, and which service?

[15:09:05] <StephenLynx> how did you install mongo?

[15:09:42] <Zelest> mawngoo deebee!

[15:09:46] <hsk3> StephenLynx: Manually. I run mongod in the background (a separate mongodb user). Then mongo can be run by any user.

[15:09:53] <hsk3> I downloaded the tarball for Debian 8

[15:10:05] <hsk3> (since there aren't debian 8 packages yet)

[15:10:39] <StephenLynx> which init system debian 8 uses?

[15:10:48] <hsk3> systemd

[15:10:58] <hsk3> i was able to set that up too

[15:11:17] <hsk3> so mongod runs on system startup

[15:11:24] <StephenLynx> then run systemctl status mongodb

[15:11:31] <StephenLynx> or mongo, can`t remember.

[15:12:11] <hsk3> ok, what am I looking for?

[15:12:31] <StephenLynx> what does it tells you?

[15:12:38] <hsk3> active (running)

[15:12:46] <StephenLynx> then you started the service itself.

[15:13:08] <StephenLynx> see how many mongod processes are running.

[15:13:29] <hsk3> how do i do that?

[15:14:07] <teprrr> ps aux|grep mongod

[15:14:10] <StephenLynx> that

[15:14:27] <hsk3> there's only one running

[15:14:27] <StephenLynx> keep in mind the grep process itself will show.\

[15:14:30] <StephenLynx> hm

[15:14:41] <hsk3> yeah two, counting the grep one

[15:14:43] <hsk3> so one :)

[15:14:43] <StephenLynx> then probably mongod ignored it when you tried to start it again.

[15:15:06] <StephenLynx> you have been using the service all the time.

[15:15:17] <hsk3> ok i'll look into it

[15:15:30] <StephenLynx> now where the config file is

[15:15:32] <StephenLynx> I forgot.

[15:15:37] <StephenLynx> I don`t do much management.

[16:15:31] <loadh> hello

[16:15:53] <loadh> I've got this issue where I've got a production DB and a DB I update using a CMS

[16:16:15] <loadh> I take a dump of the CMS DB and upload it to the production DB

[16:16:40] <loadh> the issue is that the user can save certain items locally based on the _id

[16:17:04] <loadh> but presumably, that ID can change if I recreate the data in the CMS?

[16:17:30] <loadh> should I somehow create an _id from natural keys in the data (liek the name of the thing etc.)

[16:17:39] <loadh> (if that isn't too abstract a description of the issue)

[16:18:21] <loadh> (when I say locally I mean in the native App on their device)

[16:18:43] <loadh> so basically, the issue is that the ObjectID could change...

[16:19:08] <loadh> (because I think it's based on a machine ID and a time stamp or something?)

[16:28:53] <edrocks> loadh: you can set the `_id` field when you insert a document

[16:29:53] <loadh> @edrocks > Ok, but that still doesn't sort the issue of the user's device having persisted the 'Thing' by this id which will have changed

[16:30:01] <loadh> oh wait

[16:30:04] <loadh> well yes

[16:30:09] <edrocks> you just use their id

[16:30:39] <edrocks> or add a cms id field if you want and let mongodb generate a new one for the main db

[16:31:15] <loadh> @edrocks > update DB replacing all data with new, mongo generates it's object IDs, then take a dump of that and specify _id

[16:31:29] <loadh> meanwhile, the users device has an outdated ID for that Thing

[16:31:46] <loadh> because the user fetches from the production DB

[16:32:13] <edrocks> just insert the user documents into the main db with their _id field

[16:32:26] <edrocks> use the same id between the two databases

[16:32:32] <loadh> edrocks > I'm explaining this badly

[16:32:42] <loadh> DB 1 generates the IDs say

[16:32:54] <loadh> DB 2 gets updated from DB 1

[16:33:14] <loadh> user has a reference to an object it fetched from DB 2 before the latest update

[16:33:22] <loadh> and saved it to the device disk

[16:33:40] <edrocks> also look into "offline database synchronization". I believe sql has better support for this. But I haven't had to deal with it myself

[16:33:42] <loadh> upon retrieval from the device disk, it will be inconsistent

[16:33:48] <loadh> it's not really

[16:34:15] <edrocks> are you saying you have an issue because you already deployed things?

[16:34:24] <loadh> no, but will in future

[16:34:56] <loadh> the device fetches things from DB 2 which could get all it's contents replaced by the contents of DB 1

[16:34:56] <edrocks> is db2 on the device?

[16:35:10] <loadh> the device fetches from DB 2

[16:35:22] <loadh> and stores a few things the user has 'liked' on the device

[16:35:38] <loadh> when I say 'fetches' I mean there's an API and all that

[16:36:11] <loadh> imagine that the things in the database represented articles or something

[16:36:33] <loadh> i do all my editing in DB 1, the occasionally, replace the contents of DB 2 with the contents of DB 1

[16:36:58] <loadh> the user facing API / Web Service interacts with db 2

[16:36:59] <edrocks> if db 1 and db 2 are online why not just get rid of db2 and use db1 for everything?

[16:37:25] <loadh> because I need to be able to edit things and don;t really want to not have the freedom of being able to update and verify etc.

[16:37:37] <loadh> and...

[16:37:40] <edrocks> do you not control db1?

[16:37:51] <loadh> performance wise, it's best if DB 2 is read only

[16:37:54] <loadh> i do

[16:37:56] <loadh> i control both

[16:38:29] <edrocks> have you considered setting up a replica set? you would need a 3rd node(either data bearing or an arbiter)

[16:38:40] <edrocks> https://docs.mongodb.com/manual/replication/

[16:38:43] <loadh> a who and the what now?

[16:39:07] <edrocks> it basically links your databases together. read that link

[16:39:25] <loadh> ok thanks!

[16:39:26] <edrocks> sounds like you just didn't know about it but it should solve your issue if I understand your problem correctly

[16:39:51] <loadh> sounds like it might, but also sounds like I'd need to be careful about what I put into or remove from the DB

[16:40:23] <loadh> whereas in the two DB scenario I can edit and delete and horse around with the data in DB 1 all day and decide when the time's right to take a dump into DB 2

[16:40:26] <edrocks> you need to set if you want to read from a secondary or master and your write concern

[16:40:59] <loadh> Without ever risking that the user will see dodgy data that I'm horsing around with...

[16:41:02] <edrocks> you get the choice of making your writes async or sync(acknowledged by x nodes or a majority)

[16:41:08] <loadh> ok

[16:41:12] <loadh> i'll have a read

[16:41:14] <loadh> sounds complicated

[16:41:18] <edrocks> set your read concern to master if you are worried about that

[16:41:27] <edrocks> read through that though and you should be good

[16:41:37] <loadh> awesome, thanks edrocks :)

[16:41:50] <edrocks> np, good luck

[17:22:28] <Conner05> I want to stand up a .net app using mongodb. I need to enable use authentication. Can I use .net's default auth stuff with mongo?

[17:32:55] <StephenLynx> you will have to ask .net people.

[17:51:46] <lolcat> Hello

[17:52:26] <lolcat> What uses mongodb?

[17:54:36] <lolcat> Hello

[17:54:48] <lolcat> Does it support memcomputing?

[19:48:35] <viran> noob question, I want to save the User _Id in a different document, that is the type of _id ?

[19:48:39] <viran> what*

[21:47:42] <direwolf> hey i m trying to create a auto imcrement field

[21:47:43] <direwolf> https://docs.mongodb.com/v3.0/tutorial/create-an-auto-incrementing-field/#auto-increment-counters-collection

[21:47:52] <direwolf> i m following this tutoiral

[21:48:24] <direwolf> but the find and modify query does not return value

Log file Viewer

Help | Karma | Search:

#mongodb logs for Saturday the 13th of August, 2016