Slashdot Mirror
GitHub's Website Remains Broken After a Data Storage System Failed Earlier Today (theregister.co.uk)
Github engineers are trying to repair the data storage system underpinning the code hosting website, which has been presenting users with a "What!?" error for much of the Sunday. From a report: Depending on where you are, you may have been working on some Sunday evening programming, or getting up to speed with work on a Monday morning, using resources on GitHub.com -- and possibly failing miserably as a result of the outage. From about 4pm US West Coast time on Sunday, the website has been stuttering and spluttering. Specifically, the site is still up and serving pages -- it's just intermittently serving out-of-date files, and ignoring submitted Gists, bug reports, and posts. Sometimes, it appears to be serving a read-only cache or older backup of itself, although some fresh code pushes are coming through onto the site. From the status page, it appears a data storage system died, forcing the platform's engineers to move the dot-com's files over to another box. In the meantime, some older versions of files and repos are being served to visitors and users. "We're continuing to work on migrating a data storage system in order to restore access to GitHub.com," the team said just after 5pm PT, adding in the past few minutes: "We are continuing to repair a data storage system for GitHub.com. You may see inconsistent results during this process."
INb4 "Microsoft Broke It!" - they've yet to actually take over. They're still waiting on government approval of the acquisition.
just signed the GitHub purchase.
Slashdot, fix the reply notifications... You won't get away with it...
"All ops staff required to use pure HoloLens interface with floating console and keyboard!"
*Github immediately convulses*
"There is more worth loving than we have strength to love." - Brian Jay Stanley
What is there storage back end? and what VM system? are they useing?
Does anyone remember back when they trashed most of the cloud data of Sidekick smartphone users shortly after they acquired the company? This was even covered by Slashdot!
https://hardware.slashdot.org/...
Looks like a repeat of the bumbling.
Which data storage sytem would that be? Brand and model. I want to know so that I can think badly of it, or possibly of the persons who configured it, but most probably the vendor.
When all you have is a hammer, every problem starts to look like a thumb.
A possible clue:
Moving to bare metal and federated storage on Rackspace has brought our average Rails response time to consistently under 100ms
But that post is ancient.
When all you have is a hammer, every problem starts to look like a thumb.
Another possible clue:
I’m a developer on Openstack’s object storage system, and I believe that Openstack is uniquely positioned to achieve this vision.
In contrast, I believe that Openstack is a steaming pile of impenetrable Python poo, especially its storage architecture. If this is indeed their platform then I am not at all surprised to hear that Github admins are now busy failing their files over manually.
When all you have is a hammer, every problem starts to look like a thumb.
And somehow managed to muddle through to this day without serious drama. While Sourceforge isn't to my taste, they still host a lot of projects and the price is right.
When all you have is a hammer, every problem starts to look like a thumb.
if not being able to make posts and gists breaks your workflow to the point of "failing miserably"...then you are just a miserable failure :-)
I know right? And any shining star such as yourself out to be able to still succeed brilliantly even if their source repo serves out random files.
When all you have is a hammer, every problem starts to look like a thumb.
Microsoft would just run CHKDSK to fix it.
Nice troll, that was sweet.
When all you have is a hammer, every problem starts to look like a thumb.
Thanks Microsoft.
Breaking shit and blowing stuff up...
Chas - The one, the only.
THANK GOD!!!
Nobody uses sourceforge. Outdated codebase. I forget which fork is still open source for gforge, but it's better.
For commercial projects, Atlassian is now industry standard. It's good for funded open source, too, if a little overkill.
Of course, a central server on a distributed system like git seems pointless.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
Isn't that what Microsoft did to most acquisitions?
For what it's worth, sourceforge is still here and googlecode isn't. Probably says more about Google's baby-eating culture than anything.
When all you have is a hammer, every problem starts to look like a thumb.
That thing does seem to delete files whenever it feels like.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
That's a god damn good idea! Will help all those boring days at work! Thank you kind sir for adding another troll to slashdot for the lulz.
The storage back-end is called The Cloud, and it stores The Bits. Fortunately, everything is backed up a million times in Git, so somebody has a copy.
Support my political activism on Patreon.
Git seem to be working for me. But travis isn't starting on the Pull requests.
Right. Because you'd never have more than one developer in one location working on a project. Sheesh.
Except for the github source code, which is in Subversion.
Remember, You are unique...just like everyone else.
It was only running on Subversion, but I it was perfect for a small team. And the revision discussion feature was awesome.