Portal Home > Knowledgebase > Industry Announcements > Web Hosting Main Forums > Providers and Network Outages and Updates > Faulty cable blacks out internet for millions [MERGEDx5]
Faulty cable blacks out internet for millions [MERGEDx5]
| Posted by chillipc, 01-31-2008, 09:09 AM |
|
hey all has anyone been affected by this
http://news.bbc.co.uk/1/hi/technology/7218008.stm
|
| Posted by ub3r, 01-31-2008, 09:11 AM |
|
I've been waiting for the day when a shark would chomp out part of the internet.
|
| Posted by Vinayak_Sharma, 01-31-2008, 10:37 AM |
|
Faulty cable blacks out internet for millions
Internet services badly affected in India
Internet outage hits business from Cairo to Colombo
And More
|
| Posted by chillipc, 01-31-2008, 10:40 AM |
|
yeah lol it reminds me a little like jaws 2 with cable junction lol
|
| Posted by ameeriklane, 01-31-2008, 11:14 AM |
|
The article mentions FLAG. Were any other backbone providers using that cable? I then want to look up the peers we use to see if any of those are using that backbone.
|
| Posted by smitty1258, 01-31-2008, 11:36 AM |
|
lol, one big shark
|
| Posted by eLief, 01-31-2008, 12:09 PM |
|
Well if they've been affected by that, then I don't think they will be responding
|
| Posted by Woooo, 01-31-2008, 12:15 PM |
|
i am affected, I had big trouble paying my server bills today since it was 31st.
|
| Posted by Burhan, 01-31-2008, 12:21 PM |
|
Kuwait is affected horribly. Bandwidth is up/down and tons of lost packets. Voice calls are also affected.
|
| Posted by rhenderson, 01-31-2008, 02:32 PM |
|
See the story on CNN http://www.cnn.com/2008/WORLD/meast/...rss_topstories
This took out Egypt and Pakistan among others. I guess today we do not have to worry about as many server attacks and fraud orders coming in
|
| Posted by Webdude, 01-31-2008, 02:57 PM |
|
Yeah, the terrorist organization "Spamhaus" is claiming credit for cutting the lines during their jihad against the imperialist Spammers. I personally plan on giving my support to spamhaus by sending them more scissors.
|
| Posted by Nick H, 01-31-2008, 03:02 PM |
|
Wow - what will all the hosting companies that use outsourcing do
|
| Posted by DWebby, 01-31-2008, 06:49 PM |
|
http://www6.comcast.net/news/article...ernet.Outages/
Has anybody suffered from this outage?
|
| Posted by Richard, 01-31-2008, 07:09 PM |
|
Ouch! I'm glad I don't outsource to India anymore. I would hate to try and explain to a client that an undersea "phone line" disaster is the reason we're asking for another 3-5 days.
|
| Posted by misticil, 01-31-2008, 07:19 PM |
|
That'd be pretty embarassing PR wise.
|
| Posted by drumman24, 01-31-2008, 08:18 PM |
|
LMAO!!, forget hosting companies, what about every large company out there? Dell, MS. hehe.
|
| Posted by Vinayak_Sharma, 02-01-2008, 12:06 AM |
|
For us here in India we have alternate route and providers, Flag is owned by Reliance but the main provider is VSNL and their circuit seems to be intact, yesterday morning speed was like just pathetic but today for me its back to normal.
India Will Stay Online
|
| Posted by InfiniteTech, 02-01-2008, 07:02 AM |
|
Internet communication link for hundreds of millions of people in India and and across the world were disrupted after a ship, anchoring off the Egypt's Alexandria coast, damaged two sub-marine optic cables Sea-Me-We-4 and Flag on Wednesday morning.
During this time, traffic was divert via the pacific route which caused clogs and slowdowns.
Sea-Me-We-4 is a 20,000 KM long optical submarine cable network which links 14 countries - Singapore, Malaysia, Thailand, Sri-Lanka, India, Pakistan, Bangladesh, UEA, Saudi Arabia, Italy, France, Tunisia, Algeria and Egypt.
Bharti Airtel, India's leading telecom services firm and Tata-owned VSNL are part of the consortium that owns Sea-Me-We-4 while the Reliance Group owns Flag.
As per the news paper, Times of India; both the network cables have been restored to 80%. The rest will be fixed within 24 hours. Traffic in these networks has been established.
If anyone of you had problems yesterday - don't trouble your hosts cause its out of their reach to fix this.
|
| Posted by supportfacility, 02-01-2008, 07:27 AM |
|
Yes, this is a major outage - restored 80% that is great. In another local news paper I read it may take up to 10 days.
|
| Posted by dotHostel, 02-01-2008, 07:34 AM |
|
I guess I'm in these 20% Latency from our monitoring center in Germany to a Singapore server we have at QALA increased 100ms after the incident.
Very true. When the latency increased to the same level of the other server we have at QALA -- routed via Pacific and the US -- the first thing we thinked it was a provider issue.
Last edited by dotHostel; 02-01-2008 at 07:45 AM.
|
| Posted by InfiniteTech, 02-01-2008, 07:34 AM |
|
No way 10 days. Its gonna be miserable then!
I hope while fixing this, they lay 2 more lines, I guess that will make Asia in par with Europe and the US in terms of bandwidth availability.
|
| Posted by nauae, 02-01-2008, 08:15 AM |
|
Actually this is for more serious than most people think, and as server Admins all of the WHT members should pay close attention to developments.
We have servers located in UK, Netherlands, Central and East Coast USA. TTL and response times to all of these locations were and still are dramatically affected by the added congestion on various peering links. This in turn affects how your servers and the applications/services on them operate.
For instance, check your outgoing mail queues, on all 26 of our primary outgoing SMTP servers we had and still have massive retry queues that were bring the services real close to hang-up.
DNS is not resolving for a high number of Asian IP's and the number of time outs from affected countries is resulting hung DB threads and extended error logs.
For our part we began rerouting SMTP (both incoming and outgoing) based on geographical region, and would recommend other Admins do the same.
It is estimated that it will take some 2 to 3 weeks to repair the cables and during that time traffic on the remaining links is going to be under extreme pressure. At the moment all of the primary USA based carriers are selling bandwidth to ISP's from the affected regions using Satellite up/down links - so just because you are not in the Middle East or SE Asia - don't think you will not be affected.
|
| Posted by netomatic, 02-01-2008, 12:08 PM |
|
Business has not been impacted much here, as India has multiple submarine cables landing on its shores from both Atlantic and Pacific routes.
Last edited by netomatic; 02-01-2008 at 12:14 PM.
|
| Posted by ameeriklane, 02-01-2008, 05:47 PM |
|
The bigger question is this: Why does one line outage take down so much of that part of the World? Don't they have more redundancy in place?
When I worked in the US, accidental fiber cuts were a daily occurence, but it never really affected things. Why is the ME relying on just a single cable? Doesn't that seem a bit obtuse?
|
| Posted by InfiniteTech, 02-01-2008, 05:51 PM |
|
It isn't a single cable. You address it as 'a cable' but in reality its a whole bunch of them!
What do you expect? A single optic fiber cable carry bandwidth of 14 countries in 2 continents?
|
| Posted by InfiniteTech, 02-01-2008, 05:53 PM |
|
I doubt, probably 2-3 days at worse!
|
| Posted by dotHostel, 02-01-2008, 06:11 PM |
|
Agree. However while I'm seeing small bandwidth disturbances in all monitored servers and latency increases of 10ms it is not something to worry about.
|
| Posted by nauae, 02-01-2008, 06:29 PM |
|
There are a total of 5 breaks in the two cables and although the cables are quite close to one another (400 meters apart) weather conditions alone will prevent reapirs from starting until Tuesday 5th Febraury.
Assuming they work 24 hours and the breaks are clean and the cable has not suffered additional damage due to it being dragged along by a ships anchor, then it will take 3 to 4 days for each cable.
The last time the FLAG cable got broken (just off the coast of Saudi) it took 3 weeks! And when the SeaMeWe4 got cut off Mumbai it was 2 weeks.
Wishful thinking 2 to 3 days - and anyway the cable was broken on 30th early morning - so its already been 3 days!
That part of the problem aside, routing is now being handled through SeaMeWe3 (which is around 60Gbs) and round the SE Asia coast line and across the Pacific.
Also, to address "ameeriklane" comments - the type of fibre optic cables we are talking about here are very different from the lines that criss-cross the US.
The two cables that have been cut carry 2 Terra Bits per second, the third remaining cable has a capacity of just under 60 Giga Bits per second - hence the slow down and congestion.
There are cables that run from the Middle East in an Easterl direction crossing in to the US across the Pacific - but the latency on the lines is pretty high due to the physical distance.
Obtuse maybe, but a reality for sure!
|
Add to Favourites 
Print this Article
Also Read
Shared Hosting
Shared Hosting
