View Full Version : My report on SCM systems.
TobiasTheViking
18th June 2006, 04:45 PM
This is a little (54 pages on a total of 2410 lines) on System Configuration Management, if that kind of thing interests you, check it out.
Lets hope i haven't done anything TOTALLY retarded in this report.
NOTE: presenting this as my semester project wednesday the 21st, so lets not pray that it turns out well now shall we?
Sincerely
Tobias
Report: http://tobiasserver.ath.cx/hosting/Synopsis.pdf
BenK
18th June 2006, 07:54 PM
couldn't get your link to work "Page cannot be displayed"
a_unique_person
18th June 2006, 11:21 PM
Problem with your configuration management, maybe?
TobiasTheViking
19th June 2006, 02:45 AM
couldn't get your link to work "Page cannot be displayed"
I just clicked the link, and it worked. funny.
BenK
19th June 2006, 02:54 AM
works now :cool: but it's not something that applies to me.
TobiasTheViking
19th June 2006, 03:05 AM
Hm, ok, having some problems, so maybe it was down a bit during the night,i know for a fact that it will go down a bit within the next few hours.. sorry about that.
Wudang
19th June 2006, 03:16 AM
I think a better title might be "Source code integrity management"?
jeremyp
19th June 2006, 03:59 AM
None of the systems you tested are advertised as "software configuration mamagement" systems and so I think two of your tests are unfair. I use subversion myself and I think the developers would argue that regression testing is part of some change control framework of which svn would be a tool to control source code revisions. Developers of the other systems would have similar arguments I suspect.
Furthermore, I don't know how valid your data integrity tests are. I don't know of any database system where randomly changing files within it has consistent results. The best thing you can do in such a situation is report an error and advise the admin to restore the back ups.
Finally, your file safety test is too simplistic. You need to include tests for adding deleting and renaming files and even changing the directory structure. You also need to test what happens if an update of multiple files is interrupted mid way (logical consistency). Many source code changes involve multiple files. You need to make sure that all the changes are present in the repository or none are.
TobiasTheViking
19th June 2006, 05:18 AM
None of the systems you tested are advertised as "software configuration mamagement" systems and so I think two of your tests are unfair. I use subversion myself and I think the developers would argue that regression testing is part of some change control framework of which svn would be a tool to control source code revisions. Developers of the other systems would have similar arguments I suspect.
Furthermore, I don't know how valid your data integrity tests are. I don't know of any database system where randomly changing files within it has consistent results. The best thing you can do in such a situation is report an error and advise the admin to restore the back ups.
Finally, your file safety test is too simplistic. You need to include tests for adding deleting and renaming files and even changing the directory structure. You also need to test what happens if an update of multiple files is interrupted mid way (logical consistency). Many source code changes involve multiple files. You need to make sure that all the changes are present in the repository or none are.
*sniff* you are so mean to me. :)
I'll take it into account, thanks :)
jeremyp
19th June 2006, 09:07 AM
*sniff* you are so mean to me. :)
I'll take it into account, thanks :)
Sorry.
I've just read that it's a "semester project" due to be handed in this week, so my comments are a bit late really. I'm also not sure what level of detail a "semester project" should go into, so, in that context, perhaps my comments are over the top.
TobiasTheViking
19th June 2006, 01:47 PM
Sorry.
I've just read that it's a "semester project" due to be handed in this week, so my comments are a bit late really. I'm also not sure what level of detail a "semester project" should go into, so, in that context, perhaps my comments are over the top.
No, not at all, always good with constructive critique. And you weren't mean. Anyways, this report is not what i have to hand in, i have to make a presentation, the report i just did to have some data to make my presentation from. So i can still work it into my presentation(simply stating which errors i have made, which areas i forgot to check, how it should be done, and why it should be done, will probably suffice.).
Thanks a bunch for your input. :) don't worry.
TobiasTheViking
23rd June 2006, 01:14 AM
And i got A+ :D
Raphael
23rd June 2006, 04:17 AM
And i got A+ :D
Woohoo TobiasTheKeener.:)
hodgy
23rd June 2006, 04:15 PM
How did the presentation go?
Edited - Oh, I assume you got A+ for the presentation.
TobiasTheViking
23rd June 2006, 05:24 PM
How did the presentation go?
Edited - Oh, I assume you got A+ for the presentation.
indeed
© 2001-2009, James Randi Educational Foundation. All Rights Reserved.
vBulletin® v3.7.5, Copyright ©2000-2010, Jelsoft Enterprises Ltd.