Online bugtracking and local SVN : Is it possible ?

[eanon]

Hello,

Looking for the easiest bug tracking mechanism for my needing, I have to ask for your lights.

In short, I'm an independent developer working on some projects, sometime alone and sometime with others developers. When we're several developers on a project, I'm used to use an all-in-one online project management service like assembla, github, etc. When I'm alone, until now, I've never had to use bug tracking at all (it was simple enough - tools - to do the job and manage things with flat text files called bug.txt and todo.txt)... But all changes and now, I need one (because of a bigger project)

Well, here is my current status : About a project of mine on which I'm alone on dev side, I'm close to the stage where I'll have to involve some testers (some being totally newbies in computer area, my goal being to improve ergonomy for them, specifically). Initially, I've thought to communicate with them through a simple forum (in my website, hosted by an hosting company) and install a bug tracking system locally only (on my LAN) ; so, in this case, of course, I'll have to manage bugs tracking by myself, copying/pasting from what they could tell me on forum. Not very fluid

So, ideally, it would be better if they could directly enter issues and demands by themself, in an online bug tracking system (my ISP providing Mantis beside my website). Thus, knowing this cross-platform project is locally versioned by Subversion (installed on a PC in my LAN) with several SVN clients in my development stations (one under Windows with TortoiseSVN and one under Mac OS X with SCPlugin), I would like to be able to connect my commit messages with online bugs tracking entries (i.e. be able to reach a specific Mantis entry from a link in SVN log and reciprocally). Does all of this possible ?

Also, maybe a step beyond, I'm wondering if it would be possible to install Mantis locally too, then synchronize this local Mantis with the online one. So, this way, I'll connect my SVN commits against this local Mantis only (and the two Mantis installations will sync by themself). Again, is it possible ? Do you think it's a good idea ?

I need your clarifications and advices

[atrol]

Have a look at https://github.com/mantisbt-plugins/sou ... ion#readme for integration of SVN and MantisBT.
I never tried myself, maybe MySQL database replication can be used to sync MantisBT instances http://dev.mysql.com/doc/refman/5.6/en/replication.html

[eanon]

Thanks atrol. I have some reading, now If someone else has another idea, don't hesitate. At this step, I read and listen... until the point I'll have elements (and knowledge) enough

--
EDIT : well, the question about synchronization of two Mantis seems to be not so easy. Some refs :
- http://www.mantisbt.org/wiki/doku.php/m ... issue:7970
- http://www.mantisbt.org/forums/viewtopic.php?f=3&t=82

--
EDIT #2 : maybe the solution : http://www.rubyrep.org (a two-way sync of databases) Nevertheless, it implies a question more : their example (screencast) is done between two Redmine databases, and we see they exclude two tables (one about scheme migration and another about plugins)... What's about Mantis ? What should be the tables to ignore/exclude during synchronization ?

[eanon]

up (about my question in the "EDIT #2" above)

[atrol]

Tmaybe the solution : http://www.rubyrep.org (a two-way sync of databases)

Intersting thing, but seems to be quite dead, latest commit is about two yeras ago: https://github.com/rubyrep/rubyrep

[eanon]

Intersting thing, but seems to be quite dead, latest commit is about two yeras ago: https://github.com/rubyrep/rubyrep

Effectively, but the mailing-list seems to be still active (I see messages in 2013) : https://groups.google.com/forum/?fromgr ... um/rubyrep. So, the questions could be :

1) Does rubyrep developed enough (even if not updated since long time) to do the job against Mantis's MySQL db v.2013 ?
2) Or is there a rubyrep competitors around (since, rubyrep may constitute a proof of concept) ?
3) In both cases (using rubyrep or a competitor), what are the MySQL tables to ignore in a two-way sync process ?

I've droped a quick mail to Arndt Lehmann, the rubyrep's author.