Founder and CEO of M5 Hosting here. We did have a network outage today that affected Hacker News. As with any outage, we will do an RCA and we will learn and improve as a result.<p>I'm a big fan of HN and YC in general, we host of other YC alum, and I have taken a few things through YC Startup School. During this incident, I spoke to YC personally when they called this morning.
HN is one of the few sites I always keep zoomed-in (around 200%), which led to me finding an interesting bug in Chrome while HN was down: Chrome's internal "This site can't be reached" page uses the zoom level of the site you would be visiting (if it were up), rather than Chrome's default zoom.<p>Screenshot: <a href="https://i.imgur.com/VwFtgQh.png" rel="nofollow">https://i.imgur.com/VwFtgQh.png</a>
Hacker News is hosted at M5 and they are having a network outage:<p><a href="http://status.m5hosting.com/pages/incident/5407b8e2b002442510000272/604fa9a35aef05052e955afe" rel="nofollow">http://status.m5hosting.com/pages/incident/5407b8e2b00244251...</a><p>edit: Unrelated to the Azure outage.
Had me quite confused because I'm also having home internet issues. I was trying to get my laptop to switch to my mobile hotspot and HN is one of the sites I used to test connectivity because a) it's almost always available and b) loads very quick.<p>A bit of a mindfuck trying to assess my actual internet connectivity via a site that was also down : )_
my fingers automatically just start typing in "news.y" when I'm idle, I definitely didn't know what to do when greeted with a 404!<p>Is there any way to put the HN homepage on an edge cache so at least the homepage shows up? Or am I admitting that I'm addicted to checking HN too many times a day?
I never expect HN to be down... I asked my wife - hey is the internet down? She said - no, it's working for me. I clicked on another site and my mouth dropped.
@dang<p>Would love an updated post on what the current hardware / software stack that’s running HN.<p>It’s been years since I’ve seen a post/comment on this topic.<p>Are you still running FreeBSD, on a few high frequency cores (iirc)?
Azure AAD also had an outage at this time - perhaps linked in some domino effect, or perhaps a coincidence?<p><a href="https://status.azure.com/en-us/status" rel="nofollow">https://status.azure.com/en-us/status</a>
Seeing HN unresolved was a bit weird as it is the best performing site I ever visit on my low bandwidth phone so several times I thought the problem was on my end. But in the end it helped me realize how frequently I dial into HN while it was down. I have a bit of a problem and I think I need to turn on that no procrastination flag on.
> Back now. Got to write some code for a change... - 5:39<p><a href="https://twitter.com/HNStatus/status/1371576822748487683" rel="nofollow">https://twitter.com/HNStatus/status/1371576822748487683</a>
Something that I learned from this is that HN has a status Twitter. Rarely used though, which is a testament to the team.<p><a href="https://twitter.com/HNStatus/status/1371525940656803848?s=20" rel="nofollow">https://twitter.com/HNStatus/status/1371525940656803848?s=20</a>
The one time I'm actually reading HN for actually relevant information for actual work, it's down for half a day. Made for a great excuse to take a nap.
I have a massive learning project[1], and I think 2/3 of my "to get through as soon as sensible" content is news.ycombinator links.<p>Needless to say, this site is my own personal StackOverflow, and I think there's something about ingratitude bouncing around in my mind somewhere.<p>[1]<a href="https://github.com/PhilosAccounting/ts-learning" rel="nofollow">https://github.com/PhilosAccounting/ts-learning</a>
Since I couldn't get to HN, I wrote up how to make the site resilient to outages: <a href="https://gist.github.com/peterwwillis/ce2bfaba7fc72e4af44c28135ab3db1e" rel="nofollow">https://gist.github.com/peterwwillis/ce2bfaba7fc72e4af44c281...</a><p>tl;dr 1 server x 2 providers, different regions, replicate content