5 ms·

Internet hiccups today? You're not alone. Here's why

75 points by tanglesome 12y ago

cratermoon 12y ago
Instead of spending money to prepare for this, they are working on writing new throttling plans.
elwell 12y ago
If they internet were to not work for the period of one year, what would happen? How many people would die? (A genuine question for anyone who feels up to task)
- 13throwaway 12y ago
  That would be a good question for Randell Munroe. https://what-if.xkcd.com https://what-if.xkcd.com
  - elwell 12y ago
    Submitted.
- toyg 12y ago
  s/people/careers
  - nanidin 12y ago
    It would probably launch some careers, considering someone would have to figure out what went wrong and get it all back up and running again!
- yeukhon 12y ago
  This is one of the disruptions a solar storm will bring to us. We are just lucky we weren't hit directly last time. http://www.space.com/21237-extreme-solar-storms-electrical-grids.html http://www.space.com/21237-extreme-solar-storms-electrical-g... http://www.reuters.com/article/2014/07/25/us-electricity-solarstorms-kemp-idUSKBN0FU20Q20140725 http://www.reuters.com/article/2014/07/25/us-electricity-sol... I don't think we are ready yet to embrace this possibility.
notastartup 12y ago
Just your friendly old neighbour NSA installing a new backdoor on one of the backbones, apologies for the slight hiccup you might experience, blame it on your isp
aaron987 12y ago
This explains the problem I had with LastPass earlier today. None of my computers would let me log in for about a 10 minute period.
ximeng 12y ago
Possibly related? http://noc.c4l.co.uk/2014/08/cisco-infrastructure-emergency-maintenance/ http://noc.c4l.co.uk/2014/08/cisco-infrastructure-emergency-... Following today’s impact to service, engineers have identified that the following emergency maintenance is required: Location: Cisco core Start: 2014-08-13 00:01 BST End: 2014-08-13 04:00 BST More: http://www.reddit.com/r/talesfromtechsupport/comments/2deb04/today_was_a_bad_day_and_every_isp_knew_it_was/ http://www.reddit.com/r/talesfromtechsupport/comments/2deb04... http://www.reddit.com/r/networking/comments/2dcrl2/major_network_outage/ http://www.reddit.com/r/networking/comments/2dcrl2/major_net... Package loss of 3 to 5% on a couple of Verizon routes: http://internetpulse.net/Main.aspx?xAxis=Destination&yAxis=Origin&zAxis=Metric&nAxis=Period http://internetpulse.net/Main.aspx?xAxis=Destination&yAxis=O... Active BGP entries >= 512k: http://www.cidr-report.org/cgi-bin/plota?file=%2Fvar%2Fdata%2Fbgp%2Fas2.0%2Fbgp-active.txt&descr=Active+BGP+entries+%28FIB%29&ylabel=Active+BGP+entries+%28FIB%29&range=Day&StartDate&EndDate&yrange=Auto&ymin&ymax&Width=1&Height=1&with=Step&color=auto&logscale=linear http://www.cidr-report.org/cgi-bin/plota?file=%2Fvar%2Fdata%... More discussion on reddit http://www.reddit.com/r/sysadmin/comments/2dc0qe/uk_anyone_else_seeing_major_networking_problems/ http://www.reddit.com/r/sysadmin/comments/2dc0qe/uk_anyone_e... http://www.reddit.com/r/sysadmin/comments/2dcol3/the_internet_hit_512k_bgp_routes_today_causing/ http://www.reddit.com/r/sysadmin/comments/2dcol3/the_interne... http://www.reddit.com/r/sysadmin/comments/2dcer9/anyone_else_with_level3_as_their_voip_providers/ http://www.reddit.com/r/sysadmin/comments/2dcer9/anyone_else...
legohead 12y ago
A route to one of our subdomains was not working at our hosting provider. It affected our application because we relied upon the subdomain as a cache between S3. Pages in the application would fail when cache was reloaded. Never mind our application, but there are some interesting "what ifs" out there, like if the route for some major players were to suddenly stop working...or, perhaps similar to our app, some minor routes failed but major sites relied upon them.

75 points by tanglesome 12y ago

Instead of spending money to prepare for this, they are working on writing new throttling plans.

If they internet were to not work for the period of one year, what would happen? How many people would die? (A genuine question for anyone who feels up to task)

13throwaway 12y ago
That would be a good question for Randell Munroe. https://what-if.xkcd.com https://what-if.xkcd.com
- elwell 12y ago
  Submitted.
toyg 12y ago
s/people/careers
- nanidin 12y ago
  It would probably launch some careers, considering someone would have to figure out what went wrong and get it all back up and running again!
yeukhon 12y ago
This is one of the disruptions a solar storm will bring to us. We are just lucky we weren't hit directly last time. http://www.space.com/21237-extreme-solar-storms-electrical-grids.html http://www.space.com/21237-extreme-solar-storms-electrical-g... http://www.reuters.com/article/2014/07/25/us-electricity-solarstorms-kemp-idUSKBN0FU20Q20140725 http://www.reuters.com/article/2014/07/25/us-electricity-sol... I don't think we are ready yet to embrace this possibility.

notastartup 12y ago

Just your friendly old neighbour NSA installing a new backdoor on one of the backbones, apologies for the slight hiccup you might experience, blame it on your isp

aaron987 12y ago

This explains the problem I had with LastPass earlier today. None of my computers would let me log in for about a 10 minute period.

ximeng 12y ago

Possibly related? http://noc.c4l.co.uk/2014/08/cisco-infrastructure-emergency-maintenance/ http://noc.c4l.co.uk/2014/08/cisco-infrastructure-emergency-... Following today’s impact to service, engineers have identified that the following emergency maintenance is required: Location: Cisco core Start: 2014-08-13 00:01 BST End: 2014-08-13 04:00 BST More: http://www.reddit.com/r/talesfromtechsupport/comments/2deb04/today_was_a_bad_day_and_every_isp_knew_it_was/ http://www.reddit.com/r/talesfromtechsupport/comments/2deb04... http://www.reddit.com/r/networking/comments/2dcrl2/major_network_outage/ http://www.reddit.com/r/networking/comments/2dcrl2/major_net... Package loss of 3 to 5% on a couple of Verizon routes: http://internetpulse.net/Main.aspx?xAxis=Destination&yAxis=Origin&zAxis=Metric&nAxis=Period http://internetpulse.net/Main.aspx?xAxis=Destination&yAxis=O... Active BGP entries >= 512k: http://www.cidr-report.org/cgi-bin/plota?file=%2Fvar%2Fdata%2Fbgp%2Fas2.0%2Fbgp-active.txt&descr=Active+BGP+entries+%28FIB%29&ylabel=Active+BGP+entries+%28FIB%29&range=Day&StartDate&EndDate&yrange=Auto&ymin&ymax&Width=1&Height=1&with=Step&color=auto&logscale=linear http://www.cidr-report.org/cgi-bin/plota?file=%2Fvar%2Fdata%... More discussion on reddit http://www.reddit.com/r/sysadmin/comments/2dc0qe/uk_anyone_else_seeing_major_networking_problems/ http://www.reddit.com/r/sysadmin/comments/2dc0qe/uk_anyone_e... http://www.reddit.com/r/sysadmin/comments/2dcol3/the_internet_hit_512k_bgp_routes_today_causing/ http://www.reddit.com/r/sysadmin/comments/2dcol3/the_interne... http://www.reddit.com/r/sysadmin/comments/2dcer9/anyone_else_with_level3_as_their_voip_providers/ http://www.reddit.com/r/sysadmin/comments/2dcer9/anyone_else...

legohead 12y ago

A route to one of our subdomains was not working at our hosting provider. It affected our application because we relied upon the subdomain as a cache between S3. Pages in the application would fail when cache was reloaded. Never mind our application, but there are some interesting "what ifs" out there, like if the route for some major players were to suddenly stop working...or, perhaps similar to our app, some minor routes failed but major sites relied upon them.