Stack Overflow Status

Looks like it's undergoing maintenance: https://www.stackstatus.net/

Hah tried to post this and it didn't work, glad for the confirmation. Not sure if its the right place to look, but https://stackstatus.net/ is also not working for me.

And gone again. "We are currently offline for maintenance" edit: most recent update on https://twitter.com/StackStatus is from Mar 10 - Status blog ( https://stackstatus.net/ ) is 3 weeks old

https://stackstatus.net https://twitter.com/StackStatus Neither says anything (at the time of writing).

I currently see > We are currently offline for maintenance On https://stackoverflow.com/, but I see nothing on https://stackstatus.net/

No update on their status blog as yet: http://stackstatus.net/

Even the statuspage is dead https://www.stackstatus.net/

http://stackstatus.net/ doesnt show anything (yet)

Seems to be back up. Incident page: https://www.stackstatus.net/incidents/e380b9bc-9ed2-4672-80e3-d0e090d7…

Like A site wide outage at StackOverflow being caused by an inefficient backtracking regex engine: https://stackstatus.net/post/147710624694/outage-postmortem-july-20-20…

StackOverflow also had an outage a few years ago that was caused by exponential blow-up of a backtracking regular expression. https://stackstatus.net/post/147710624694/outage-postmortem-july-20-20…

Probably something mundane like ^[\s\u200c]+|[\s\u200c]+$ That's the one that took down Stack Overflow a few years ago https://stackstatus.net/post/147710624694/outage-postmortem-july-20-20…

DoS with regex is a thing: https://www.owasp.org/index.php/Regular_expression_Denial_of_Service_-… StackOverflow had a similar case a while back: https://stackstatus.net/post/147710624694/outage-postmortem-july-20-20…

Thanks for adding. As a related aside, someone who read the post pointed me in the direction of http://stackstatus.net/post/147710624694/outage-postmortem-july-20-201… Apparently Stack Overflow had a 34 minute outage this past year due to user input that resulted in backtracking. What's crazy is how innocent the regex seemed, too.

Indeed. To support backreferences, "regex" libraries are forced to use algorithms that can be very slow in the worst case. The sad thing is that the libraries use the same algorithms even if the expression doesn't contain backreferences. A while ago, Stack Overflow had a brief outage because of regular expression performance, although the expression that caused it didn't even use backreferences or other non-regular features: http://stackstatus.net/post/147710624694/outage-postmortem-july-20-201… In contrast, Google Code Search - when it still existed - supported regular expression searches over world's public codebases. One key ingredient making this possible was to only use proper regular expressions: https://swtch.com/~rsc/regexp/regexp4.html

http://davidvgalbraith.com/how-i-fixed-atom/ http://stackstatus.net/post/147710624694/outage-postmortem-july-20-201… https://swtch.com/~rsc/regexp/regexp3.html The difference isn't academical. It has huge practical implications, from software that is just annoyingly slow to DoS attacks. People need to understand that. And stop using non-regular regexes (or better yet: regex in general) in production code. It increases world-suck.

This http://stackstatus.net/post/115305251014/outage-postmortem-march-31-20… :)

$&%^ maybe they have been hacked!!!! OMG!!! Being a little more serious. They have actually had some sort of outage as described on there website. http://stackstatus.net/post/85902805739/stack-exchange-primary-data-ce… Stackexchange network status updates are on there twitter feed. https://twitter.com/StackStatus

It's a thing: https://en.wikipedia.org/wiki/ReDoS And it really happens. StackOverflow had a ~30 minute outage a couple years ago because of it: http://stackstatus.net/post/147710624694/outage-postmortem-july-20-201… One of the key points here is that it's not always obvious from looking at a regex whether it will exhibit catastrophic backtracking (at least, to an untrained eye). Even more insidious, catastrophic backtracking may only happen on certain inputs, which means that tests won't necessarily track it either. Therefore, desiring a performance guarantee here is certainly not "strange." The bottom line is that the various implementation strategies for regex engines have trade offs. On Internet forums, it's popular to "proclaim" for one side or the other. The backtracking advocates like to believe that catastrophic cases never happen or are easily avoided and the linear time advocates like to believe that so long as matching doesn't take exponential time, then it will be fast enough. :-)