open All Channels
seplocked EVE General Discussion
blankseplocked Possible Connectivity Issues with Tranquility [Updated 21:39 UTC]
 
This thread is older than 90 days and has been locked due to inactivity.


 
Pages: 1 [2]

Author Topic

CCP Fallout

Posted - 2011.02.02 21:57:00 - [31]
 

Originally by: Sliona Snafe
Problem is External? Does that sugar coat your marketing image by saying that?


No, but we can't exactly control LINX or other providers. Well, I do have the mighty powerful fez... hrm :)

The Post Man
Posted - 2011.02.02 22:11:00 - [32]
 

Originally by: CCP Fallout
Originally by: Sliona Snafe
Problem is External? Does that sugar coat your marketing image by saying that?


No, but we can't exactly control LINX or other providers. Well, I do have the mighty powerful fez... hrm :)


You sure it's LINX? Wouldn't that **** the whole internet for all of the UK?

I know several Brits who aren't having any trouble with Eve (Interestingly almost all are with Virgin Media) and I've personally had no trouble with the internet, just with Eve...

Phorneus
Amarr
Yulai Guard 1st Fleet
Yulai Federation
Posted - 2011.02.02 22:15:00 - [33]
 

Sometimes I need to close my client and reopen it like 3 times to work around a strange login issue...

Filodar
Posted - 2011.02.02 22:16:00 - [34]
 

Originally by: The Post Man

You sure it's LINX? Wouldn't that **** the whole internet for all of the UK?

I know several Brits who aren't having any trouble with Eve (Interestingly almost all are with Virgin Media) and I've personally had no trouble with the internet, just with Eve...


99% of the stuff you do online can deal with a couple of seconds of delay and TCP connections getting reset so you wouldnt notice.

Also it did affect most of the UK see https://stats.linx.net/cgi-pub/exchange?log=combined.bits&png=daily

LINX has 2 separate LANs some people are just lucky their traffic wasnt going over the one which had issues.

Interestingly I know several people in corp who are on Virgin and it did mess up for :)

Robdon
Vanishing Point.
The Initiative.
Posted - 2011.02.02 22:21:00 - [35]
 

Yeah, it wasn't Eve related.

Was funny, both my eve sessions worked fine, but my TS just wouldn't connect, said the IP was unreachable for about 20 minutes.

CCP Explorer

Posted - 2011.02.02 22:58:00 - [36]
 

Originally by: Akira Katsurou
do something to jita please, srsly, lagging really bad, i cant load sell orders
Can you please provide more detail? CPU load on the Jita location node has been very reasonable tonight and there are no queues on the Forge market node.

This could be an effect of the network issues that have been in London tonight, e.g., your client could be having problems receiving the reply from the server.

Durin Sarga
Posted - 2011.02.03 00:08:00 - [37]
 

EVE gate connection to mail, calendar, having issues. FYI

Siabhra
Amarr
Interstellar Archaeology
Posted - 2011.02.03 01:26:00 - [38]
 

Edited by: Siabhra on 03/02/2011 01:27:33
Before I can even get to the login page, I get an error that says my "Eve installation may have modified, damaged, or corrupt files. Please run repair.exe located in the EVE directory or re-install the game client".

WAAAAHHH!

I could not find a repair.exe file. Can someone advise as to where that would be? I am running Windows 7 and when I go to my programs and open the EVE folder I can not see this. ACK!

Help?

Pexan
Posted - 2011.02.03 02:49:00 - [39]
 

Over the last 3-4 days I have been mostly experiencing 5-15 second periods of lag/disconnection, but then opening the wallet seems to "jar" something loose, the connection returns - very annoying but playable if you are in a system alone.

Last night and tonight I have been getting client disconnects/crashes - getting worse. I thought my ISP could be the problem, being I'm sort of in an "outpost town" in Colorado, but everything else is working fine (albeit slow compared to most of you).

I just crashed again, 7:45pm MST

Katell
Posted - 2011.02.03 07:59:00 - [40]
 

Originally by: CCP Fallout
It appears that the issue is not within our network and is external.


With all due respect:) Isn't that a false statement?

Surely the internet is your network and if we don't manage to connect to it you need to help us troubleshoot and make alternate solutions so we can connect.

anywho~ I think you [CCP] are doing a great job administering the network, except yesterday I've never experienced any problems with your game (except desynch lag which doesn't really count because it never is logged anywhere^^)

keep up the good work, happy customer 3 accounts.

Robdon
Vanishing Point.
The Initiative.
Posted - 2011.02.03 08:43:00 - [41]
 

Originally by: Katell
Originally by: CCP Fallout
It appears that the issue is not within our network and is external.


With all due respect:) Isn't that a false statement?

Surely the internet is your network and if we don't manage to connect to it you need to help us troubleshoot and make alternate solutions so we can connect.



It was a problem with the LINX network... ie the network that connects the UK to the internet. How would CCP help with this? There isnt really anything that they or you can do when something as 'central' as this goes wrong.

http://en.wikipedia.org/wiki/London_Internet_Exchange

Louis deGuerre
Gallente
Malevolence.
Posted - 2011.02.03 13:50:00 - [42]
 

Originally by: Katell
...Surely the internet is your network...


*facepalm*

Othran
Brutor Tribe
Posted - 2011.02.03 15:58:00 - [43]
 

If you're Aussie and on Optus then the LINX problems are not the issue for you - your issue predates that and seems to be in the Tata network which Optus use for international traffic. ie you need to hassle Optus.

Othran
Brutor Tribe
Posted - 2011.02.03 16:08:00 - [44]
 

Originally by: The Post Man

You sure it's LINX? Wouldn't that **** the whole internet for all of the UK?


No.

There are two public peering LANs within LINX.

Brocade and Extreme.

If your ISP is on the Brocade LAN (Be, Virgin and a load of others) then you'd have had issues. On Extreme then you'd be largely unaffected for routing to Tranquility, but you'd still be having major issues with some other traffic.

There's also private peering (private interconnect - another 700Gbit/sec), plus there are other peering points (LONAP, PacketExchange etc) so even if something drastic happens it shouldn't (in theory) bugger things up much. In practice of course there really isn't that much slack.

Izun
Caldari
Best Path Inc.
IT Alliance
Posted - 2011.02.06 15:32:00 - [45]
 

Edited by: Izun on 06/02/2011 15:38:01
Sadly, there's still issues.
I know it's not your fault, but for sure you'll be able to get more information than us.
This is my tracert.

Edit: I live in Italy. tracert done at 16:30 CET.

Cynosural Love
Posted - 2011.02.08 10:11:00 - [46]
 

As members will have noticed, there have been three incidents in the last week on our Brocade-based LAN. I would like to bring you up to date on what we are doing about the instability that started in November 2010.

Firstly, Mike Hellers and I visited Brocade HQ on 27th & 28th January 2011,and this was definitely a confidence restoring experience, notwithstanding the subsequent incidents.

I'll briefly cover the three incidents that occurred, and then say what we are doing to restore a stable platform [first notification times on ops-announce given in square brackets]:

1st incident [02/02/11 06:09]: this was low light level triggering the known 'low lag utilisation' bug. We have tested and verified the link, and the cause was a faulty XFP/transceiver, which has been replaced.

2nd incident [02/02/11 20:42]: this is believed to be a Traffic Manager bug, 8x10G card RMA requested

3rd incident [03/02/11 15:27]: we saw Traffic Manager errors affecting a limited number of members on one 8x10G card. We moved all members on that card to different ports on SW16. The card is currently still in the switch, as Brocade asked us to find ways to generate some traffic on that card for them to further analyse the issue. We have asked for a detailed plan of what they want us to do and what they plan to do during those tests

In relation to the first incident, and ever since we have known one of the ways that this particular issue is triggered, we have been working on improvements to how we monitor adverse optical performance in our networks.Since last Monday, we are capturing optic levels from all ports which are part of ISLs once per day. This is not yet fully integrated into our alarm monitoring (which will happen later); but even in its current state, it will help us to detect potential issues with faulty/degrading fibres or transceivers.

There have been a spate of Traffic Manager related issues, and all the known ones are scheduled to be fixed in the 5.000e software currently in Brocade QA. We have agreed a plan with Brocade to get all the MLX switches onto more stable, bug-fixed code, and we are working closely with them to minimise the risk of that process. Included in this are two rounds of off-network testing at LINX, using additional hardware that is being shipped to us for this specific purpose. Brocade intend to release this code at the beginning of March, and we will apply it straight away.

Among the many other hardware, software, monitoring and process related improvements that have also been agreed with Brocade, we have done the following:

* improved the periodic ping system between all of our monitoring systems across all member ports. This was put in place some weeks ago following analysis of various previous incidents, initially as an hourly check, now increased in frequency to every 15 minutes;

* implemented a new command script, developed last week and deployed last Friday, using a new debug command given to us by Brocade;

* enabling LACP on all of the ISLs. We have already enabled it on two ISLs, and will gradually add more over the coming days/weeks. Since a LAG group reconfiguration is required, we will enable LACP whenever we have to take down an ISL for maintenance or other reasons. Any remaining ISL which hasn't been touched until we upgrade to the 5.000e software will be changed during that maintenance. This will also help to prevent these issues in a much more reliable way - and which should also prevent incidents like the first one noted here from happening again;

More generally still, I give you my assurance that we are taking all prudent steps to address these continuing problems - and that we are considering all possibilities (short, medium and long term) to get back to
the operation of a stable exchange platform.

Cynosural Love
Posted - 2011.02.08 10:14:00 - [47]
 

We intend to devote a full session to talking about our network development plans at LINX72 on 21/22 February 2011, and a senior Brocade executive (Ken Cheng, Vice President, Service Provider Products) will be attending and presenting at that meeting. We have made provision for an extensive Q&A session at the close of day one of the meeting.

As ever, I am quite happy to have an individual dialogue with members if you have additional questions that are not addressed in this update.

John
--
John Souter, CEO, London Internet Exchange Ltd


^^ Email I have been privy to

Veebring Greetings
Posted - 2011.02.08 11:27:00 - [48]
 

Looks like the UK is implementing new surveillance measures.

Ay Liz
Sacred Templars
RED.OverLord
Posted - 2011.02.09 18:48:00 - [49]
 

I need to necro this as there are still random disconnects for a bunch of people. I suspect it is within Telia and Level3 this time.
Theres also this thread on Issues, Workaround and Localization -> http://www.eveonline.com/ingameboard.asp?a=topic&threadID=1462176

It is frustrating..


Pages: 1 [2]

This thread is older than 90 days and has been locked due to inactivity.


 


The new forums are live

Please adjust your bookmarks to https://forums.eveonline.com

These forums are archived and read-only