All the cool kids post code to Github these days, and many open source projects have adopted Github as their main tool for managing the development of their projects. Just recently it was announced that Python is moving their main repository to Github. For the last few hours (nearly 6 hours), Github has been struggling with a major network outage. This was also just noted by The Register. Hopefully Github's engineers will get to the bottom of it. At least with Git, even if Github is down, developers can still work with their code and commit changes to their local repositories and then push later when service is restored. This is but one example of the pitfalls of cloud hosting, though I daresay that Github's uptime is much better than most IT departments' own servers.
Last time similar things happened (Mar 2015), it was blamed on China.
[NOTE: The above was submitted shortly after 1:00 AM (UTC) on 2016-01-28.
GitHub has a status page which, as of this writing, is reporting: "Everything operating normally."
GitHub also posts a list of recent status messages. -Ed.]
(Score: 3, Insightful) by PizzaRollPlinkett on Thursday January 28 2016, @12:00PM
So a distributed source code control system has a single point of failure? The whole point of the Internet was to allow it to continue working if any one part failed. Now people can't seem to get away from centralized services. These central services allow people to do things that weren't possible before, but seem to be introducing single points of failure. If a few services like Gmail, Facebook, and GitHub went down, the entire Internet would stop functioning. Is this the right kind of progress?
(E-mail me if you want a pizza roll!)
(Score: 0) by Anonymous Coward on Thursday January 28 2016, @01:10PM
Oh noes, the "like" buttons on our pages went away! When those sites are turned off, it will hardly be the end of the Internet. Now if you'd said Cloud Flare, I wouldn't argue with that.
(Score: 0) by Anonymous Coward on Thursday January 28 2016, @03:04PM
*TRIGGERED*
(Score: 2, Touché) by redneckmother on Thursday January 28 2016, @05:33PM
What are these "like buttons" of which you speak?
:)
Mas cerveza por favor.
(Score: 3, Informative) by tibman on Thursday January 28 2016, @03:22PM
Github isn't git, it just uses it. Git isn't like svn or tfs or other version control systems. Every git client is also a server. If github (or your own central git server) goes down then you can still push code to each other (or anywhere). People can even push code straight to your local machine. Think of it like your webbrowser is also a webserver and you have your own local copy of the internet. If the real internet went down you'd just use your local copy. To keep it updated you could pull from other content/website creators directly. Or if you made changes you could push those directly to other people. Having a central place only makes it easier for everyone to collaborate.
SN won't survive on lurkers alone. Write comments.
(Score: 0) by Anonymous Coward on Thursday January 28 2016, @03:36PM
Even in a system like that it is advantageous sometimes to call someone 'master'. That is what github does in a nice easy way. I can still pull/push as a master from my local copy into others. But github is a nice way to have a master that is 'always there'.
(Score: 3, Informative) by darkfeline on Thursday January 28 2016, @11:37PM
Most people use Git as a centralized VCS. The concept of distributed VCS is beyond most people, just like the concept of functional programming, although JavaScript has helped alleviate that somewhat.
Personally, I have Git repos all over the place: laptop, USB, GitHub, Bitbucket, private Gogs server. It allows me a lot of flexibility in working on code (example: GitHub down? Just push/pull from Bitbucket instead. No Internet? USB to the rescue) and allows me to easily handle very complex merge operations that would make a CVS user commit genocide.
Join the SDF Public Access UNIX System today!