While SO is down (and other sites) they link to the ServerFault blog. That's fine with me. However what if all sites go down (what just happened). The link to ServerFault´s blog also doesn't work anymore then. Perhaps it would be nice to link to something else then?

Or just remove the link to the blog all together.

| improve this question | | | | |
  • Actually, SF's blog was up during the outage. – Kevin Jan 28 '12 at 19:21
  • 3
    @Kevin No it wasn't I got an error page. Or was it a glitch that just happened to be be at the same time as the maintenance outage? – PeeHaa Jan 28 '12 at 19:22
  • 2
    I clicked on the link from the "We're down" page and got it just fine. – Kevin Jan 28 '12 at 19:23
  • 1
    I didn't get the page – PeeHaa Jan 28 '12 at 19:24

We had to pull down the blogs, the latest post had a couple of very large images in it (this will be fixed!) as well as our offline pages not pulling from the CDN yet. The traffic was saturating our internet uplink which caused our remote people to not be able to work over the VPN as well as timeouts etc for some people on the offline pages - if you've ever tried remote connections over a slow or saturated link you'll know what that is like. You can get things done, but well it is SLLLOOWWWWWW.

Network Utilization during outage

| improve this answer | | | | |
  • 7
    So, the important takeaways are: a) Stack Exchange employees can work from home, and b) they are forced to work on weekends. Awesome =) – casperOne Jan 28 '12 at 19:39
  • Takeaway - @casperOne never worked for an employer where every minute of downtime was costing $$$ ;) – Oded Jan 28 '12 at 19:43
  • 4
    @casperOne Re a), I'm quite glad I don't have to make the commute from Berlin to New York every morning. The trans-atlantic bridges are always a bottleneck. – balpha Jan 28 '12 at 22:03

Funny story - we were planing on having the blog up, however there were a few images that were quite larger than they should have been. This ended up saturating our upstream bandwidth and we pulled it.

offline death

| improve this answer | | | | |
  • 3
    Tsk tsk. Images that are not web optimized? On the SO blog? – Oded Jan 28 '12 at 19:44
  • @Oded SF blog, not SO blog – Jeff Atwood Jan 28 '12 at 19:53
  • 2
    @JeffAtwood - Much worse then... – Oded Jan 28 '12 at 19:54
  • 2
    I'm more surprised that VPN traffic shares the same uplink as web-traffic for this very reason. – MDMarra Jan 28 '12 at 23:13
  • Sorry about that, guys -- apparently the status screenshot I took was a whopping 600k, which was pretty big. I take responsibility for this one; I only hope it results in a gold badge of shame. – Peter Grace Jan 29 '12 at 1:40
  • @mdmarra its due to the way the colo charges for bandwith, one large file transfer could easily double our bill – Zypher Jan 29 '12 at 2:13
  • @Zypher interesting. I'll have to bug you or Pete about this in chat on Monday. – MDMarra Jan 29 '12 at 2:19

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .