Silkroad Online Forums

Suprise!

The issue is not Joymax, @Home, Adelphia, Time Warner, Charter, or Comcast's fault.

As many of you know, SRO is located in Korea. What has happened is that Korea has updated its international routing tables. This has left most providers without access to the proper records and left searching for the host. There are two ways to prove this.

1. http://www.traceroute.org - Run a trace from any US router listed there to either of SRO's IP addresses. The trace will die at a random router within Korea. There are some of those traces that will complete. The reason is that the change to the tables has propogated to their network.

2. As you can see with most of the in network traces here, most users are dying before even getting off their ISP's backbone. The reason for this is because of the way routing is set up. When you send your packets from your cable, dsl, dial up modem they must pass through a series of routers either due to OSPF or static routing. If using static, there is a definite path known and the packet will continue until a destination is unreachable. With OSPF, or open shortest path first, routing the destination must broadcast the location of the host in order for the packet to find its destination. If no path is broadcast the packet hops along the backbone until it gives up and "times out".


Also, for those who believe that you cannot access SRO due to DNS or reverse DNS issues, you honestly do not understand DNS. All reverse DNS does is points a mail server to its host. What on earth do you care if SRO's mail server is pointing to its host for? You all received misinformation from your ISPs with that regard, however, they are all correct in telling you the issue is on SROs end. Though in a way, it isn't on their end either. I hope this makes sense to everyone.

hmm, u kno abit about networking eh? haha , so do u happen to kno if they will try to fix it or what?

The problem is that there is nothing to fix. Whoever maintains the Korean network needs to make Level 1 providers aware of the change and allow access so that the new equipment may begin to ARP up. At this moment, I believe the issue is within the Korean network(possibly why the game has not updated you?). A trace from Level 3, Telus, and Datapipe show that they are making it into Korea and then have no clue where to go from there. The only thing you can do is try to get more info out of the game. I apologize, however, I do not play the game or I would do so myself.

hm, not sure of what i should do, can we just wait it out, or will they ever be notified of this, or what?

To be honest I can't really give you any good information on that. Things like this usually happen very quickly within the US, however, I have no experience within the Korean Intranet. I would imagine things should propogate fairly quickly though, and if not, they need some new admins!

to me 2 weeks isnt very fast, hope they do something about it soon

To be honest I am not sure of the exact date that the change was made. I too would be upset if I was unable to get to it as well.


As for those who bashed service representitives for thier ISPs, you should be ashamed of yourselves. First you blow an issue rediculously out of proportion claiming you know the problem, when in fact you had no clue what you were talking about. Honestly, its a shame because those people put up with your crap daily, when they usually know more than yourself.

Well, if you would be so kind as to help us out. do you know any possible way we could inform a Korean server of this etc. Or Is it possible that it may never be fixed?

Well if it would be because of DNS i could still be able to open http://121.128.133.11 or http://121.128.133.12, could't i?

Yes, however, most people complaining of the issue have not been able to surf via IP address either. If you can, what are the DNS servers you are currently using?

I use my work (university) server's dns, whose cache replicates thru ISP LITNET dns, which is a part of GEANT network, that is broadcasted via Telia in Europe and so on (Britain, USA, Korea)... Lots of routers to travel thru and looks like attempts to connect fail at Korean servers, not at any server before.

Wow, I may as well just have my friend's dad adopt me for a couple months till TBC comes out...I'd love to stay and try to force Koreans to do something for us americans....BUT, I'll pass. Hopefully the korean's that control the internet arent' anything like the rest of the government...

well if this is true, then why is only aldephia/tw having the problem? if this was the problem, then no one would be able to connect, and can you explain how homestead(dutch ppl) got there problem fixed?

No offense, but can you read? There are multiple ISPs having problems. Go the link I posted in my first post and run a trace from any of those domains. I can almost guarentee it will not work. I am at my girlfriends college currently and she is unable to access the site either. There is no Adelphia or Time Warner in this area. Just a little proof before we get more speculators.


C:\Documents and Settings\owner>tracert http://www.silkroadonline.com
Unable to resolve target system name http://www.silkroadonline.com.

C:\Documents and Settings\owner>nslookup http://www.silkroadonline.com
DNS request timed out.
timeout was 2 seconds.
*** Can't find server name for address 192.168.157.6: Timed out
Server: ns1.ppltelcomip.com
Address: 209.183.191.162

*** ns1.ppltelcomip.com can't find http://www.silkroadonline.com: Non-existent domain

C:\Documents and Settings\owner>ping 121.128.133.11

Pinging 121.128.133.11 with 32 bytes of data:

Request timed out.
Request timed out.
Request timed out.
Request timed out.

Ping statistics for 121.128.133.11:
Packets: Sent = 4, Received = 0, Lost = 4 (100% loss),


NOW, FOR THE REAL PROOF THAT WE ARE NOT ON ADELPHIA/TW:

C:\Documents and Settings\owner>ipconfig

Windows IP Configuration


Ethernet adapter Local Area Connection:

Connection-specific DNS Suffix . : HERCOLLEGE.edu
IP Address. . . . . . . . . . . . : 10.20.239.67
Subnet Mask . . . . . . . . . . . : 255.255.255.0
Default Gateway . . . . . . . . . : 10.20.239.1


A 10.x.x.x is a private IP address that is not routed to residential customers. It is routed from the DHCP server at my girlfriends college.

You will notice an nslookup of that IP address will not resolve from any ISP.



Ok, lets forget about this connection for a bit and travel off network to some other domains and run traces.


Traceroute Output
FROM customers.datapipe.net TO 121.128.133.11.

traceroute to 121.128.133.11 (121.128.133.11) from 66.70.0.11, 64 hops max, 40 byte packets
1 vl405.dist1-1.ewr.datapipe.net (64.106.131.210) 0.388 ms 0.385 ms 0.364 ms
2 core2 (64.106.128.18) 0.521 ms 0.555 ms 0.371 ms
3 GigabitEthernet5-0.GW1.EWR17.ALTER.NET (157.130.254.161) 0.424 ms 0.345 ms 0.551 ms
4 0.so-1-1-0.XL2.EWR6.ALTER.NET (152.63.28.89) 0.571 ms 0.480 ms 0.569 ms
5 0.so-3-3-0.XL2.SEA1.ALTER.NET (152.63.104.41) 82.074 ms 95.846 ms 82.031 ms
6 194.ATM7-0.GW7.SEA1.ALTER.NET (152.63.105.181) 74.160 ms 74.101 ms 74.081 ms
7 koreatelecom-gw.customer.alter.net (208.214.97.30) 74.302 ms 74.360 ms 74.685 ms
8 220.73.154.17 (220.73.154.17) 200.652 ms 200.582 ms 200.518 ms
9 apnc11 (211.216.216.101) 215.986 ms 232.897 ms 215.997 ms
10 59.18.52.25 (59.18.52.25) 228.680 ms 228.860 ms 228.706 ms
11 220.73.149.2 (220.73.149.2) 212.908 ms 212.914 ms 213.183 ms
12 211.192.60.38 (211.192.60.38) 212.062 ms 211.905 ms 211.817 ms
13 * * *
14 * * *
15 * * *



Type escape sequence to abort.
Tracing the route to 121.128.133.11

1 toroonxngr00.bb.telus.com (154.11.63.85) [AS 852] 0 msec 0 msec 0 msec
2 154.11.10.78 [AS 852] 72 msec 72 msec 76 msec
3 KoreaTelecom.PLALCA01GR00.bb.telus.com (154.11.3.246) [AS 852] 76 msec 72 msec 76 msec
4 211.48.63.181 [AS 4766] 204 msec 204 msec 204 msec
5 218.145.63.217 [AS 4766] 204 msec 204 msec 204 msec
6 220.73.150.57 [AS 4766] 204 msec 204 msec 204 msec
7 220.73.148.2 [AS 4766] 204 msec 204 msec 204 msec
8 211.192.60.6 [AS 4766] 204 msec 376 msec 232 msec
9 * * *
10 * * *
11 * * *



Both traces get well into Korea and then all the sudden stop. Why would this be? Could it be that everything I told you was correct? Or is this a mass conspiracy against every company in the US?

I should hope you are smart enough to choose the correct answer now.

Also for Gob, the reason why some people can gain access and others haven't is because the Level 1 providers that the US travels through have not been made aware of the changes to the routing table. Other than this game, think of the last time you visited a website hosted in Korea. Now, think about how often your mom or dad might? Now think of your grandmother. This is not a common route, and therefore, low priority. They will wait until the change propogates through most networks until it is fixed.

Also, lets point out that Time Warner and Adelphia both have their own backbone solutions. None of thier equipment has been purged together yet. So if you are looking for a link between the current merger between the companies you will not find one. You can tell which backbone you are currently on by doing a traceout to any random website. If you are on Adelphia's old backbone it will read as "adelphia_com". If you have been migrated to the TW network, it will read as "TWC_core".

And also, I would be interested in hearing what the Dutch ISP did to fix the problem. Possibly the Dutch made contact and updated the reflected change to thier routing tables?

I suppose I won't know, unless we have an insider on this page.

So we can't complain to Adelphia that they suck for not being able to connect us to SRO while most other ISPs had no problem at all?

First, Adelphia doesn't exist. Atleast not as an ISP. You now either belong to Time Warner or Comcast. So your problem is with one of them. However, both companies are using the Adelphia backbone(now owned by TW) until all customers have been migrated to thier new networks.


Second, suppose Adelphia could contact Joymax. What on earth would that do? Is anyone here reading what I am telling you? The problem is not on Joymax's end either. The Koreans have reset thier routing table. Go to the website provided, find another provider, run the trace. Those providers cannot access either.

Lets go through this one more time...

YOUR ISP IS NOT RESPONSIBLE FOR YOUR ACCESS TO ANY WEBSITE, UNLESS THE PROBLEM IS THIER OWN FAULT, SUCH AS DNS.

As stated, this has absolutely nothing to do with DNS. Hence, not your ISPs issue. Honestly, it is not worth my time if you guys cannot read and understand simple sentences.

You aren't that Allen guy are you?

---------------------


Even if it's not Adelphia/TW/Comcast's fault, they should try to help us so we can get onto SRO because we PAY (or our parent's do for those not old enough). But NO! They keep saying it's not their fault. Ok, we get that. Now do they care to help us? If not, let's just switch to another ISP like AOL, because so far, no one from AOL has complained about not being able to connect. You don't seem to acknowledge that other ISPs don't have this problem. Why the hell can AOL costumers connect? Because they provide better service?

lol if it is allen i wouldnt complain, i would of gone here and been like
STFU STOP MESSAGING ME ABOUT SHIT U DONT KNOW Farking NOOBS
OH AND ADELPHIA HATES JOOOO

But yea WTF can we do? Call the Korean Government that runs their shit and say uh well um alot of us europeans and americans cannot connect to your WONDERFUL games so can you put it back the way it was or fix it?

I don't know about you, but I really hope Southern Korea controls the government that controls the internet.

But alittlebird, what people really just want to know is:
What can we do to fix/help speed up/notify that there is a problem to someone? Most people that couldn't connect just gave up...what really can we do? Anyone here know Korean?

wow you can get that far on trace?? man i can get up to 4 then it times out .. this is BS!!!!!!!!!!!!!!!!

1. I have no idea who Allen is. As stated, I do not play this game.

2. Hitman...The only thing you can do is contact the maker of the game. They will need to contact thier own providers.

3. c5, I sincerely hope you are kidding. Did you read any of my posts? Those traces are from other providers listed on http://www.traceroute.org

At some point, your connection would hit those providers and you would continue your trace to the destination, that is if the providers knew where that was. Look at the traces, they both come from a different host, Datapipe and Telus. How would I trace from two different hosts on one location?

I cannot trace from this location as my girlfriends college network blocks all ICMP and un-needed UDP traffic. For those that do not understand simple networking, providers block simple traffic that is not needed in order to mantain stability. Pings and traces are useless on a network such as this, due to the fact that the student takes it or leaves it and the network is for "reasearch" as opposed to gaming and high impact downloading/surfing.

As for the reason you cannot get past "hop 4" on your traces, please refer to my first post. TW, Adelphia, Comcast(and most other ISPs) set thier backbone up with OSPF routing. If the backbone cannot determine where the host file is located in the world it will continue to hop back and forth on the backbone until you eventually see a time out due to a non response to your machine. To better understand that, realize that a ping is just an echo, timing how long it takes a packet to get to its destination and it then reports back and tells you the speed. If the packet does not return, you have no reported speed.

Now why do you die at hop 4 as opposed to the hops previously?

The hops previous to 4 are either not set up to route or are not routers and do not have the ability to tell a packet where to go. Your first hop could be one of two things. If you have a router, the first hop is your router(192.168.x.x). If there is no router, the first up is your local CMTS or headend(the cable plant you get your immediate connection from). From there, it hits other pieces of equipment on a "static route" or a predetermined route. Here is the scenario:

hop 1 - Your local cable plant inspects the packet and makes sure it is allowed on the network, it cares not where the packets destination is. It sends it on to the next hop, albeit a router, edge router, or backbone. No matter what kind of packet this is, it will travel the same path. No matter where the packet is going, it will travel the same path. On to hop 2.

hop 2 - Acts just as hop 1 unless it is the backbone.

Continue this same path until it hits the backbone, the actual routers. From there, your path to the host is clearly defined. If a path has not been broadcast on to the backbone, the packet will be disgarded and die a horrible dropped packet death.

In the case of the SRO issue, the routing table is not matched up and the path is not being broadcast to multiple ISPs. The backbone is dropping your packets because of the wasted effort.

I hope this makes a bit more sense.