As a new wannabe-user I just installed and updated to RockStor version 3.9.1-1.
While I cannot compare improvements with previous versions (as I just came in on the current one) and really appreciate the effort of all concerned, I do want to see if we can change the priorities of things on the TODO list of the maintainers a bit.
I have, from my brief use so far, the following problem. The commercial value and useability of RockStor is severely destroyed by the bad state of maintenance and updates of the basic parts of RockStor.
Examples ???
AFS (Netatalk) is version 3.1.7 dates back to november 2014 (there are 4 versions after it, most recent march 2017)
Docker is version 1.13.1 we are now at 1.17.6 but admittedly they seem to have jumped from v 1.13 to 1.17 (strange)
Owncloud docker app is also not up to date at all, (could not get it to run so abandoned it, this is not good!).
These are the core apps that EVERY user will want to use and rely on in their daily business environment and the main reason why we/I/my-customers might choose RockStor. And (now I am a bit harsh, I know…) while being able to use disk-encryption is a “nice-to-have”, it is IMPERATIVE - if not just for security reasons most surely - that ALL these basic modules are kept uptodate and are not lagging behind so many versions.
I and my (and thus your) customers are fighting with issues, security risks and problems that are resolved already long time ago. But not for the RockStor user because of the maintenance neglect.
And please do understand that I appreciate the effort of all maintainers that spend time and effort to improve RockStor. I and my customers are more than willing to pay the maintenance fee even if it was 3 times more expensive, heck maybe even 5 times. In its current state, RockStor is NOT the superduper NAS solution it could be and should be. I want that changed. I want RockStor to rock and sell itself. There is however some effort required to get it there… But can we PLEASE get the priorities right?
It would be a nice feature to have a more customizable front end for rock-ons, advanced tab maybe? Where you could have a “Portainer like” front end to allow users to install/manage there custom images or select there own dockerhub images to install. You could then scale back including every request for a rock-on and putting it in the rock-on repository while giving the users freedom to select say there own Plex docker repo.
Why not just include portainer? It’s open source, it seems to address many of the issues the custom rockstor interface is handling, and works with all docker images.
I didn’t know about portainer. Having had a quick view at the site, I do like the idea. It seems like a very developed product so if that were incorporated into Rockstor, then that would free a lot of developer time. Certainly the maintenance of the rock-ons would also not be required anymore. So spending that time and a little TLC to the main setup of Rockstor, one could have an extremely useful new version of Rockstor. As I said before, the basics are there, it just needs some serious polishing.
I do not wish to meddle into the Rockstor development team, but I do believe they should consider this option. I for one said before that I would welcome Rockstor becoming the killer program that I think it could be.
If the team needs someone to write documentation, I could free some time to do that, if need be.
I’ll have a go at trying portainer. Looks like fun.
Regards
Luc De Meyer
PS…
I set up a server with a minimal CentOS7, samba installed with 2x 3TB raid1 btrfs sets giving 5.5 TB Raid1-secured storage for datastore, configure some shares and tested it to have an average of 120M/s transfer rate over the net from a windows10 client (GREAT result for my purposes).
I then installed docker from the repo and then activated the portainer app.
Very interesting that this went flawless and fast.
It is now too late in the evening and I have a lot of docs to go through now. But very happy with the result so far. The management screens from the portainer app look very good. I hope I can find my way quickly. A good acid-test.
Hello,
I think the rockon’s are perfectly fine. Compared to portainer the installation of rockons is simpler and can target a bigger audience + the docker settings are well tested to work in Rockstor enviroment.
Regarding the upkeep of the rockons it more or less the community that submit them and i think it’s a good approach as each individual can say there need in the forum and get help by as good as anyone to get the rockon added (with developers approval that is).
It has been brought up a few times that the rockon interface needs a face-lift for the ever growing repository so i’m sure it’s on there backlog but it’s summer time and we all need a break
With that said i do wish there was an easier way for people with “less” skill to test and help validate rockons. Perhaps a local repository in the rockstor interface where someone can import a rockon .json file and also delete ofc. I think that could help to speed up development of the rockon side of things.
Hi all,
short answer: some “Rockstor packages” (pls read long answer to understand these quotes) are outdated probably because of an old kernel.
long answer: Rockstor is a Centos distro, so asa Centos repos update you get newer packages, but Rockstor kernel is not released with Centos repos (we use to test new kernel with btrfs), so probably everything depending on new kernels will be a little outdated
Thanks Mirko, that makes sense. Slow change is the nature of CentOS, both for better and for worse.
What are your thoughts about Portainer? It seems to be a much more general-purpose tool for docker management than rockons, and doesn’t seem to be significantly more complex to configure.
Hi @phren0logy,
never had Portainer, but it seems cool. Rock-ons is a Rockstor ad-hoc tool fully integrated with Rockstor WebUI, Portainer seems a stand alone tool - please consider licensing too, all Rockstor parts are under GPL, don’t know about Portainer
