suggestion: extend the project leadership

[hippi777]

currently the development seems to be very slow, there are a lotsa pending pull requests and very small bugs in the codebase. so my simple suggestion would be to give privileges for one or more trusted developer(s).

i'd like to know if you, @serghey-rodin, are in lack of time; if it's temporary only; or willing to do so; or whatever should we think about the future? i don't know if that is a free feature of github to make a project with more leaders (i'm still only a basic github user), but it would prevent to put main development of vesta to a fork, and so that would make it a stronger competitor in the area.

i'm interested in other's opinion as well.

i'm not an applicant for this task anyway, as i only got a job to customize vesta, mostly the appearance, so that's a private development, and i can't give that back to the community, because that's beyond my range, and that job is about to be done soon, and i'm in lack of time, just like the most developers...

btw i like vesta, it's kinda well organized with satisfying feature set, even if it have a bunch of small issues here and there, so thx and cheers for everyone around! :)

[PatrickePatate]

I agree, that's a good idea !

[n1trux]

For me, the slow development and merging of (even security-critical!) bugfixes is a huge problem. The solution would be developers with more time and I doubt there are such people right now.

We could just form a temporary (!) fork of this, fix all the bugs and @serghey-rodin can merge the whole thing back.

Would be nice if somebody could speak out who already has a working fork, else I'd probably do it with a bottle of Jack Daniels and a super long free weekend :/

My main problem would be creating a testing environment, since vestacp is geared towards bare-metal installation and an install inside a VM or Docker seems to need a serious rewrite.

see also: #1215

[hippi777]

i think @serghey-rodin dind't get lost forever, because most likely he would mention it and give away leadership, but most likely he have some holiday time, except if something bad happened to him... so we can go forward with a temporary fork, as he didn't check in for more weeks now, and would be nice to move forward, but one day he will come back, and then we can ask him what should we think...

i still dind't check out the installer, but probably a different access point is all that's needed for the fork, and for testing, a ~10G? partition can be enough with dualboot, if you would really take this role... probably you have a server for the case if the installer would use vestacp.com instead of the git repo, and i hope there is no additional magic out of the git repo, i think i'll take a look for this in a hour, if i've got anything valuable, i'll come back with it...

i'm fresh in the world of vesta, i'm not even a host provider, but a developer with a kinda fresh job to customize a lotsa stuffs around; and i'm in lack of time like most all the developers, so i would't take this quest for myself. this week we will deploy with my boss what i have made, and after that i won't be abled to use updates, but only fix stuffs by hand (or maybe a local git could help me to merge important bits without overwrite my work, but that's my interest :D ) so i don't know how deeply i will be involved into the world of vesta later, but i'm interested in the depths of the install process, so there's a chance, that i'll be abled to stay active, and help around the temporary fork, but this is mostly upon my boss...

and thx anyway :)

[peterbrinck]

As I've said before in the issue @n1trux is referencing, I would be happy to help work on a fork of Vesta, even though that I, as well, doesn't have that much time, but I use Vesta on a few client projects, so it would be nice to get some updates. And since it IS open-source, I think it should be all okay. Isn't it one of the main reasons to make anything open-source in the first place, to let the community work together and/or create a new project around it?

We've seen this with NextCloud (fork of OwnCloud) and Gitea (fork of Gogs). There's more, I just can't remember them at the moment.

Both where projects that some people thought could do better with a new way of doing things And I think, we may be in a similar situation with Vesta

[n1trux]

i still dind't check out the installer, but probably a different access point is all that's needed for the fork, and for testing, a ~10G? partition can be enough with dualboot

You can also just create a VirtualBox VM. For testing you'd need a local DNS or edit your hosts file though.

@peterbrinck would you like to create a new repository and merge some of the old PRs, then test? If not, I'll do it on the weekend and add you as contributor. Maybe we could already find a "to-do list" of uncomplicated PRs to merge and test?

[n1trux]

I have brought my fork to the current master version and will commence a clean-up this evening (delete branches; rename some stuff, edit readme). My aim is to test pull requests this weekend.

https://github.com/n1trux/hestia

@peterbrinck, @szbnwer ok with you working on this with me? I will provide QA/testing and managing merges only; but since we already have a lot of code sitting around in the PRs I think that's okay for now.

To be clear: This should in no way split the community and I do not want to create a new official fork. The aim is to be compatible to the main vesta repository to merge back in the future, kind of like a testing distribution of it.

[hippi777]

i can't promise anything, i will cooperate if i will, but don't rely on me, sorry! i wrote above what's my case, so i really don't know how much i will work in the future around vesta, but currently i have deeper interest in my job, then in vesta, as i'm not a host provider, but i wanna keep a good job, and that takes most of my focus, and i've got really no much free time... after deploying my current work (probably this week) i think i will only maintain the existing system, and get other tasks, my boss even wanna get a fully new app from the ground, that he have said will be started around december...

however i've read the bigger half of the installer (for debian) and i've seen it's pulling stuffs all around in small pieces from the site via wget, but it could use a git package for everything without getting everything from the web, so it could become portable this way, only the install and the update scripts should be modified for this...

the faster way to go is to modify the site link in the installer, so that can solve kinda everything if all the contents are on their place on a new server...

and for the community it's surely good to keep it all together, but i think the best is to make it installable without anything else but the git package, it would be even a bit more fast and nice, but i think wait for Serghey to say something is necessary anyway, maybe he will accept a modified version, but if he wouldn't accept that, then only the installer and upgrade scripts shouldn't be merged back...

the community would surely break, because of the legacy systems out there, but if we won't get help then a fork with a better background could go very soon very far... i think it's mostly Serghey's interest to prevent this, as it have some business background for him currently, that is not protected, but if i've seen that well, those 2 extensions are in the code behind a simple check, and the code is under gpl, so it's really in he's interest - hope that he have some good times on a beach and he will come back soon with good news :)

btw in worse case you should know that the docs are on the page which is under copyright, so if Serghey won't come up with any good news in time, or any time, then those should be reproduced from the ground, as that's already not a derivated work so a new fork can have it's own documentation...

[hippi777]

the installer is even vulnerable as it uses http and there's no redirection to https, it's even not a good practice to download and run anything, as that is proven to being vulnerable, even if only by the provider...

there is also (the absence of) the missing letsencrypt.pipe if it won't appear later in the code, but i've seen that in our installation...

it could take a deeper review anyhow...

[peterbrinck]

Sorry for the late response @n1trux and @szbnwer

I totally agree with you, @n1trux, regarding not splitting the community. But as I see it, the development is so slow, we have to like shake their grounds a bit, I don't know.

Anyways.. I'm more than willing to help! I have a free weekend (at least as of right now), so I can do some work most of the time. Though, I don't have any deeper understanding of the installation process and the internal operations tying Vesta to nginx, Apache etc. I do know a fair bit about the software itself separated from Vesta.

@n1trux let me know if you need anything. I'll look through the different pull requests and feature requests later today, and see if something's urgent to get going

[sonaht]

Hello guys, I am not a contributor as such (would like too anyhow), but I am a user on this cp. For what is worth, I believe that this is a great project and deserves to be continued and probably extended as well. I would also suggest that more people (in due time) must be enabled to merge and "unstick" the project. A fork (in any way, shape or form) is not always a good idea, but is not a bad one either.

[n1trux]

seems that the development has continued somehow?

I'll continue cleaning up the installation scripts and then try to get my fork to the point where we can easily create PRs for this repository.

[anton-reutov]

@n1trux

seems that the development has continued somehow?

Yes The development did not stop

[JOduMonT]

Hi guys; I like you're initiative and look forward; my concern is your fork might become permanent

When I'm looking at https://github.com/serghey-rodin/vesta/graphs/contributors @serghey-rodin did'nt code a lot in VestaCP for the last few months (close a year) + it seams to not responding since few months (since march) now.

However; my question; do you really believe it's suitable for prod ?

[dpeca]

I already have 10 production servers on VestaCP. But I must say that all servers are on Debian8 (if you wonder why - see who is second contributor on link you provided - I fixed a lot of things on Debian), and I'm using default instalation (nginx + apache + mysql + vsftpd). CentOS7 is also fine supported. BUT - all other variants - VestaCP on Ubuntu - or - VestaCP with php-fpm - have bugs.

And also - I'm using latest code from github - but from my fork ( https://github.com/dpeca/vesta ) - to avoid bad commits that are not tested (but somehow pushed to origin-master) - I'm merging new commits from origin-master only when I see that commit is OK.

Also - it is recommended that you have experience with server administration, PHP and Bash - just in case something unexpected occur - so you can quickly fix it.

So, answer is YES - it is suitable for production, but under the conditions that I mentioned.

[n1trux]

my concern is your fork might become permanent

It's not intended to be permanent and it's already incompatible, a test bed so to say.

[JOduMonT]

@dpeca I will just pull you're leg a little bit but the time you wrote it and now
now you are the 3rd not the 2nd.

anyway I'm glad you raise you're hand and very thankful for your advices

from them; I already learn I'm a black swan by running exclusively VestaCP NGINX PHP7 MariaDB on CentOS7 with https://github.com/SS88UK/VestaCP-Server-Installer

[dpeca]

Glad to see @anton-reutov fixed templates :)

It's not impossible to run php-fpm on VestaCP, I just said there are bugs (or there WERE bugs, or there are bugs on Ubuntu), probably small bugs and easy to fix for admin with expierence :) Unfortunatelly I never put php-fpm on servers, so I didn't have opportunity to go through those bugs, otherwise I'll fix them. I hope I'll have opportunity and time in future to take a look at Vesta's php-fpm stack.