I think we need (I say "we" because anyone can help fixing issues) a bug tracker system to track the bugs, and help with the fix, verify, etc.
In this way we will have a more easy scenario to work. I want to help fixing issues with ZM, but the forum method to post the bugs I think is no adecuate.
We have available a few GPL systems, like:
Bugzilla: http://www.bugzilla.org
Mantis: http://www.mantisbt.org
Eventum: http://eventum.mysql.org (I have involved with this project)
Scarab: http://scarab.tigris.org
Etraxis: http://etraxis.sourceforge.net
and so on...
Tell me what you think. ZM need more hands to work with, and with a bug tracker the work will be organized.
Edit: I missed to say we need (if not exist yet) a SVN repository.
Bug tracker request for ZoneMinder project
Logs
Sorta mentioned this before on another post, but that one was more comprehensive, and this one could evolve into that one eventually.
I wouldn't mind (and probably have time soon) to toss together a quick shell script that will spit out the relevant logs to most questions, cat the results out in the end, which the user than can copy/paste into the forums. First question from any answerer always to be "show me your logs", and the next response from the questioner is "which logs, where are they"
I'm thinking a quick find (to locate the actual logs), then a tail -20 (or whatever amount of lines would be necessary) from each log.
Could probably even bring up the result in a nice HTML window, complete with code tags around the logs. If I get around to getting it working, we could put a "prepare debug email" option on the main screen, or at least a script to execute.
Any interest? We seem to spend an awful lot of time on the board reasking the same questions, and the average newbie hasn't a clue to check the log for "shmget" to then check the faq for it. Would make the new users happy faster, and the old-timers annoyed slower.
I wouldn't mind (and probably have time soon) to toss together a quick shell script that will spit out the relevant logs to most questions, cat the results out in the end, which the user than can copy/paste into the forums. First question from any answerer always to be "show me your logs", and the next response from the questioner is "which logs, where are they"
I'm thinking a quick find (to locate the actual logs), then a tail -20 (or whatever amount of lines would be necessary) from each log.
Could probably even bring up the result in a nice HTML window, complete with code tags around the logs. If I get around to getting it working, we could put a "prepare debug email" option on the main screen, or at least a script to execute.
Any interest? We seem to spend an awful lot of time on the board reasking the same questions, and the average newbie hasn't a clue to check the log for "shmget" to then check the faq for it. Would make the new users happy faster, and the old-timers annoyed slower.
Well, this is the first project I see who hide the bugtracker and svn repo. Why? For one or two who abused the system, and all the community who want help, we can't for those two or three bad boys.cordel wrote:SVN and the bugtracker exist but they are not publicly available likely as they tend to be abused.
I don't mean SVN commits will be open, but only for read only.
Related to the bug tracker I can't understand why it is closed if you can restrict access (write access) with a registration (plus captcha) system.
I hope these rules will be reviewed for the project owners. A lot of guys (like me) want to help in a more dinamically proccess.
I believe both SVN and bug system are on private machines is one reason they are not publicly available. Another reason is most people fail to search or do any research before posting, I had a bug tracker available on a project for 6 months and spent as much time sorting and removing/merging duplicate issues as I did on the forum answering the same questions which in turn are usually again duplicates.
Phil does a fine job of tracking issues whether they are posted on the forum or emailed, but alas in the end it is his call and when he is available he will comment.
Phil does a fine job of tracking issues whether they are posted on the forum or emailed, but alas in the end it is his call and when he is available he will comment.
Which is the whole reason why the FAQ is there, It even states which logs are relevant and what to look for, yet it seems that many don't know better to check there first. Some of that is even duplicated in the rules troubleshooting post, yet again most of the posters do not even bother to research so unfortunately all this effort while it sounds good would still likely go to waste.coke wrote: First question from any answerer always to be "show me your logs", and the next response from the questioner is "which logs, where are they"
We seem to spend an awful lot of time on the board reasking the same questions, and the average newbie hasn't a clue to check the log for "shmget" to then check the faq for it. Would make the new users happy faster, and the old-timers annoyed slower.
You are right. Also me publish duplicate postscordel wrote:Another reason is most people fail to search or do any research before posting, I had a bug tracker available on a project for 6 months and spent as much time sorting and removing/merging duplicate issues as I did on the forum answering the same questions which in turn are usually again duplicates.
Anyway I think there is a good reason to make public the SVN repo for read only access, and gain access to the bug tracker.
Maybe Bugzilla is very dificult for newbies to search, but Mantis can show in the main page the last x issues to avoid duplicate bugs.
Exactly my point. Noone checks the FAQ (guilty myself in the beginning). I know they should, but aren't you sick of telling people to check the FAQ? If there were a "diagnose" button of some sort, it could in addition to collecting relevant information, open up the FAQ in another window so maybe they could try it themselves before posting.
I'll make one in my spare time (not sure when that is), and you/phil can toss it in as a patch if you like, or it can just live in user contribution forum, though noone would ever find it there, since they don't search posts first..
I'll make one in my spare time (not sure when that is), and you/phil can toss it in as a patch if you like, or it can just live in user contribution forum, though noone would ever find it there, since they don't search posts first.
.Which is the whole reason why the FAQ is there, It even states which logs are relevant and what to look for, yet it seems that many don't know better to check there first. Some of that is even duplicated in the rules troubleshooting post, yet again most of the posters do not even bother to research so unfortunately all this effort while it sounds good would still likely go to waste.
Coke you welcome to make one if you like, I'd just concentrate directly on the script it self and be sure to include the distro that it works for and hopefully maybe with some combined effort we can make something that works across all the distros.
As for a bug tracker, I can put one up and we can see how that works out. But I suspect that it will likely end up being more work than it's worth.
As for a bug tracker, I can put one up and we can see how that works out. But I suspect that it will likely end up being more work than it's worth.
Thank you Cordel!! I will make the effort to instruct the people to use only the bug tracker to repot the bugs or posible bugs, and search before report.cordel wrote:As for a bug tracker, I can put one up and we can see how that works out. But I suspect that it will likely end up being more work than it's worth.
-
zoneminder
- Site Admin
- Posts: 5215
- Joined: Wed Jul 09, 2003 2:07 pm
- Location: Bristol, UK
- Contact:
The main reasons the svn repo is not accessible is (a) I don't really know how to do it safely and as it's on my home system I don't want to opoen up any risks and (b) there are development branches that contain commercial or NDA'd information and I don't know how to exclude certain branches from publishing.
I use bugzilla for bugtracking but again that is on my home system (as I use it for other non-ZM projects) and I don't want my internet swamped by external accesses. If I can find a simple way of mirroring it onto zoneminder.com I certainly don't mind opening that up.
I use bugzilla for bugtracking but again that is on my home system (as I use it for other non-ZM projects) and I don't want my internet swamped by external accesses. If I can find a simple way of mirroring it onto zoneminder.com I certainly don't mind opening that up.
Phil
Just to add my 2c to this one:
I've been using ZM now for a couple of months and one of the things I keep thinking of is how to improve what is already an awesomely great system. But without a public svn/bugtracking system it feels like any work I may do on this would just be left unused, or just used by me. Or I may be duplicating something that's already being worked on.
Would it be too big a pain to port the current code to something like sf.net or maybe google's version of sf? That would provide an svn/cvs system together with a bugtracking system that is being used by most opensource software all over the world. I understand that it may make things a bit more sluggish for Phil, but then it should only be when checking in/out code, and should not really have much of an impact in the code/compile/test phases.
Maybe one of us asking for a public svn/bugtrack could setup project pages at sf/google and then Phil can try that out to see how it works for him? I think that would not have much of an impact in anything NDA/custom that Phil is working on since that would still be hosted by him on his own machines?
I've been using ZM now for a couple of months and one of the things I keep thinking of is how to improve what is already an awesomely great system. But without a public svn/bugtracking system it feels like any work I may do on this would just be left unused, or just used by me. Or I may be duplicating something that's already being worked on.
Would it be too big a pain to port the current code to something like sf.net or maybe google's version of sf? That would provide an svn/cvs system together with a bugtracking system that is being used by most opensource software all over the world. I understand that it may make things a bit more sluggish for Phil, but then it should only be when checking in/out code, and should not really have much of an impact in the code/compile/test phases.
Maybe one of us asking for a public svn/bugtrack could setup project pages at sf/google and then Phil can try that out to see how it works for him? I think that would not have much of an impact in anything NDA/custom that Phil is working on since that would still be hosted by him on his own machines?
I don't see any way to be able to keep two svn's in sync while omitting branches from the other in a reliable way. Attempting to keep two different svns in sync would likely just be asking for trouble.there are development branches that contain commercial or NDA'd information
Also, that really changes nothing for submitting code, As you would only have read access anyway, there would be no commit unless you are an accepted developer. You would still need to email in your patches or post them on a bug tracker.
I am setting up a bug tracker but ran into a small issue with my hosting provider otherwise it would be up now, Should be done soonish but if it gets abused it will come down.
Cordel, how bug tracker can be abused?cordel wrote:but if it gets abused it will come down.
I suppose you have restric free report, or anonymous report, only for registered reporters.
If you choose a group of good members for help and moderate in the bug tracker (like duplicated bugs, or not valid bugs) I think there is no problem, and the bug tracker never will be abused.
Humild I can offer me to help or moderate the bug tracker.