just added

[macbishop]

I don't understand anything.
It's wasting time to update script for vhcs yet isn't possible to update rc3 from rc2, we need to reinstall from 0.

This is an embarrassment, It has too many rewritten code and changed too many things, when in fact it was said that only would eliminate the bugs for the first stable version.

People can not be reinstating full every month and each time there is less monitoring of the bugs, is a naive to think that there is only 1 bug (now only 1 ticket error, 3 of vhcs script)
--
querer es poder

[joximu]

Hi macbishop
There is the "big" update script (ticket 55) still open - RatS is working on it.
When this script has been finished, then RC3 of ispcp could be released (more or less). Then, for RC4 it's planned to only make the changes for the different distributions and maybe some bugs and 2 weeks later (according to the roadmap) we can launch th ispcp 1.0 party.
My little change I did (and wrote about in post #1 here) is only if you do a manual update from an earlier version (not tooooooo old) - it helps to detect missing configuration variables.


Well, in autumn 2006 I also got the idea that ispcp (vhcs omega) is "only" a better (or finished) vhcs 2.4.8 - and then it's planned to build a new control panel from scratch because the fundament is not that great to build up a really flexible and modular panel.

We'll see...

Joximu

[raphael]

Quote:If we want to be one of the best cps out there.

we are long away from being a *real* CP

Quote:But! you didnt have a testing maschine. So every change wasnt tested before. Therefor we had a bunch of new bugs If your code would be tested before its commited, then I personaly think its ok .

ack, the main problem was that I've only got to reach isp-control.net's testing server once or twice.

Quote:Dont say, the development isnt going further because of the update script. Ask yourself why you dont help!

did you really read what I said on my post?

Quote:So raphael can spend his time to new features and the other can bring RC3 to the end

an other branch isn't a solution, the problem is that RC1 shouldn't have been released at all; it was bogus, there were plans to make major changes but they still decided to release RC1.

A simple and far better checking script would be this:

Code:

#!/bin/sh

file2Check=/etc/ispcp/ispcp.conf

if [ ! -z "$1" ] && [ -f "$1" ]; then
        file2Check="$1"
fi

[ ! -f "$file2Check" ] && echo "$file2Check couldn't be found!" && exit 1
[ ! -f configs/ispcp.conf ] && echo "configs/ispcp.conf couldn't be found!" && exit 1

VARS=$(egrep "^[a-zA-Z0-9_-]+( )?=" configs/ispcp.conf | cut -d= -f1)

for var in $VARS; do
        if ! egrep "^$var" "$file2Check" > /dev/null; then
                echo "$var was not found in $file2Check"
        fi
done

some test results:

Quote:CMD_CHMOD was not found in configs/dists/redhat/ispcp.conf
CMD_SHELL was not found in configs/dists/redhat/ispcp.conf
APACHE_CUSTOM_SITES_CONFIG_DIR was not found in configs/dists/redhat/ispcp.conf
BACKUP_ISPCP was not found in configs/dists/redhat/ispcp.conf
BACKUP_DOMAINS was not found in configs/dists/redhat/ispcp.conf
DEBUG was not found in configs/dists/redhat/ispcp.conf

Quote:CMD_CHMOD was not found in configs/dists/ubuntu/ispcp.conf
CMD_SHELL was not found in configs/dists/ubuntu/ispcp.conf
APACHE_CUSTOM_SITES_CONFIG_DIR was not found in configs/dists/ubuntu/ispcp.conf

... and those checks should be made at make time (on the 'test' target of course)

[Zothos]

Quote:

Quote:Dont say, the development isnt going further because of the update script. Ask yourself why you dont help!

did you really read what I said on my post?

In my first paragraph i have spoken to you, but the rest was not only for your ears/eys

Quote:Seriously I mean, if no one is going to take care of all the mess and problems I'll consider rewriting many parts from scratch and make it a separate project. The kind of development ispCP is going though is NOT possible for a project of this kind! this isn't a toy for someone to play with, is it?

What do you want to do raphael? A fork of a fork?

[joximu]

@raphael:

Quote:... and those checks should be made at make time (on the 'test' target of course)

well, I agree. It'd be better to check the things one at the beginning. I also like bash scripts :-)
Until now the installations does not cover the situations with an already installed ispcp. In my eyes the ispcp-setup script could also do some checks and also handle an already installed ispcp (e.g. not abort if anything is not as expected...).
Would be really great...
But hey - I don't know what the grwat Updatescript of Rats will do - so I don't spend time on this only afterwards to see that RatS is doing the same...

There are so many little and not so little tasks but also many ways to do it...

@Zothos:

Quote:What do you want to do raphael? A fork of a fork?

why not?

/J

[gOOvER]

Why not a fork of a fork? I think it's a nice idea.

Zothos, with what should we help. In the Past, when someone ask to help, they are ignored. And to test the Updatescript? No i don't test this. The guys where cry for this script should test it. And if no one will test it, we don't need this script.

[Zothos]

No, not realy.....

The guys who need a update script, are mostly the people out there with less *nix experiance.... So they dont know how to test all the stuff either....
Also, they cant upgrade manualy because of the missing skills.
I understand why you have this opinion, but is a bit selfish!

Most time, people ask to help. And you tell them what they how they can help, most of them dont do anything... Just take the phantastico stuff for example.
If someone realy want to help, then dont wait until someone tells you what you can do. Do something. As i said in my last posts...
Even if you just verify or reverify the bugs on our bug tracker. It would be a huge help! And thats not the only thing everyone can do! Improving the docs is another example. There are so many thing everyone can do without the need to ask everytime....

[mata]

a fork of a fork is not a solution either, you will still have the same shitty vhcs base.

Some people here tought they could use the downfall of vhcs to start this .. i dont blame em for trying .. cause alex was an ass. but it will have the same flaws as vhcs IMHO.

It is again a one man project.. and any effort i put into regulating, ruling moderators, or any kinda structure proposals where easely wimped off because they are to busy with 1.0 and it has to wait until finished.. but in the meantime we get a community wich is again NOT a democracy....

I will still moderate the forums .. but i simply do not believe that with all the probs going on .. ispcp is gonna release something production quality .. not in the first year.

I am not the person who is around most, but because of that i can look at it more globally then some of you, only bickering about 1 ticket ..

cmon guys you are almost there .. a lil water in the wine and we'll get a step closer to 1.0

[macbishop]

Totally agree with Mata.

I said a long time ago...

http://www.isp-control.net/forum/not-mor...t-916.html
--
querer es poder

[bpratt]

BeNe Wrote:What about a RC4 Branch ?
So raphael can spend his time to new features and the other can bring RC3 to the end. The time we loose so is not more so much.

If RC3 isn't going to make it full release, then I agree that you need to get RC3 out the door to the masses, and start work on a RC4. As that way you will have a whole lot more people up to a common version where they won't be whinging about ancient bugs from RC2c anymore !


Once you have people up to RC3, then you remind them that the nightly builds are available and can be done without too many problems.


I think that I am like many others who started with RC2c, and thought that RC3 was very close by, so did not bother with the incremental nightly updates, with the thinking of "why bother when RC3 is just around the corner", and now with such a huge manual job to update their RC2c to a current nightly build, they are less likely to spend the huge amount of time to do that.


I also think we all know that RC3 is not going to be the one before Version 1.0 , so perhaps it really is time to get RC3 out the door as it is, with a complete listing of the change requirements for those on RC2c (that's if there's no update script available).

That way you guys can get a start on RC4, and in the mean time you'll get a lot more people up to much more recent and a whole lot more relevant version of RC3.

I speak for myself, but I would not be totally surprised if many others agree with me, when I say I don't really care if we end up doing a couple more Release Candidates before Version 1.0 is released.

I think having long delays between Release Candidates is really hurting not only the development, but the credibility of this project, as it appears from an outsiders perspective to be a stagnant project. People need to see things happening, and to the public, RC2c was the latest thing !