Hello guys,
I'm trying to set up a bioinformatics hub in the style of arch linux. I's starting with the wiki, but the install is having issues when I try to get it to complete the web setup and commit the configuration. I'd also like someone I could just ask if something else comes up. For the obvious reasons, I have to rely on security through obscurity at the moment, sorry.
Sorry I can't help with MediaWiki but have you considered DokuWiki? Its very easy to set up, has all the features you're likely to need, easy to edit and maintain, etc
I doubt direct parody of features. What's the pro/con?
I'd love to help you do this. Where are you at with it now any changes? I've spun up a vm and am installing on arch now. It sounds like a configuration issue. Are there any logging outputs?
You may not want to give out your IP. I would get a .tk address.
It all resolves to the same IP trivially enough, so I'm not going to worry about it now. I need to set up the repository for packages and get content documentation.
I like your attitude about the IP I feel the same. Unless your trying to advertise the IP is good enough.
This looks cool. No https? Perhaps as a todo?
I like to run services like these behind a reverse proxy. It acts kinda like a firewall for your web (and similar) services. As you may know already, the request may or may not require the domain name to resolve the desired page depending on configuration. I let the initial proxy handle all the SSL stuff. Some frameworks, take some work to make this happen though.
Right now its in early early development, so only basic security is what I'm worrying about. So, maria no root, ssh no root, medium length passwords, and http with a standard as possible installation. The reverse proxy seems like an interesting concept, however.
I need artwork for the logo right now :P
you are right about rebuilding their wiki.
https://wiki.archlinux.org/index.php/ArchWiki:About
"
Goals
The primary objective of ArchWiki is to provide the Arch Linux user community with the most comprehensive and the most accessible documentation on the web. This is a world-wide project with many participants from all corners of the globe. The wiki embraces The Arch Way with respect to simplicity. ArchWiki is constantly evolving and thrives on collaboration; there is always room for improvement.
"
Everything they do over there is brilliant!
Right? And when I look at the hodge-podge that is the bioinformatics community, I can't help but think "you guys have no idea what you're doing, do you?". It's really bad.
LOL, those guys do need some help. Post your progress, I'm interested. :)
In the labs page, I'm compiling basic contact information. I'm going to be sending out an email to most of the world's labs. That should get a few things rolling. Granted, I want the AUR equivalent running, and a URL through my university before I send that one out. I'll update as I go along.
Ok, I put a little work into mimicking your installation.
I got a working ssl config, touched on letsencrypt for the ssl cert control, setup site logging in nginx (which can later be used for data analysis, awstats is what i am used to) we will see if we can tweak the logging to meet that, or look for something new. We will also build for reverse proxy, so that awstats (and any other admin related stuff), can be hidden behind the same server requests, using domain names (we can use hosts file for testing and the domain name for live deployment.) If you'd like to get going with out the details, I think that is the path we need to take, I'll drop the details on here in the next couple days as I find time.
oh, note to self, we need a log rotate too.
All stuff I need to learn more about :P
What I'm doing now, actually to help stability and help with the goal of zero downtime is split out the VMs as follows: wiki, bugs, repo, userrepo, database, and users, with a prototype copy to be used later to act as a backing image. I'm thinking I'll network database connections to the database VM from wiki, repo, and userrepo. All VMs will reply on users for login info. This should be a more complex setup, but much more fault tolerant, and in the case of a hack or move things can keep running.
At some point, I'm thinking I'll make the systems globally R/O where possible to keep things open. I'd love to have the systems secure enough to have global read on the entire system and have them still secure. In the mean time, once I get this working want copies of what I'm running?