Yeah I call it how I see it. 10 years ago the issues would alraedy been fixed by now
That’s a cool story. Both North American servers are working okay for most people since I still see tons on in town. So that tells me it’s most likely outside their network. Many ISPs share routing partners, but I’m not interested in convincing anyone.
Edit: For anyone who comes across this, I recommend running the looking glass to see if it shows where the issue is occurring.
Have a good night!
2 Likes
During the COVID-19 pandemic, most service providers are struggling to accommodate work from home, various other activities which have fully saturated some content delivery networks (resulting in bad routes / hops, dropped packets, a whole host of other issues, not just lag in general.)
If you are having A LOT of problems, my suggestion is to get a separate DNS stub resolver like DNSCrypt-Proxy, the latest version at this point in time being version 2.1.1 (don’t use your service provider’s DNS if this is a big hassle.) I would also get a VPN, and then ONLY connect your VPN to a server which is very close to the one you want to play on. For example, if it was a US-West realm / server you would want to select a VPN server from Los Angeles. I use EDNS Client-Subnet (Quad9) with a blank address, in conjunction with the ODoH protocol to fetch my DNS records for a stub resolver (it’s VERY close to having a full-blown caching-only DNS name server, such as one you could configure with BIND / Unbound DNS / PowerDNS / etc,) which in turn is fed directly to my VPN, so I have a reliable source of answer records.
I have zero issues / problems, unless the servers are actually down, I can still connect. This is because I can avoid bad routes / hops, not only because my EDNS Client-Subnet gives me a direct connection to the DNS server when I need to resolve an address, but also because I can physically open a socket as close as possible to the nearest server, via the WireGuard tunneling interface, which helps me to avoid avoid all that congested traffic in between (if you don’t live in the US, or you live quite a ways away from the server you play on, then this is not a bad idea.) That’s ONE way to do it, but not the ONLY way.
It isn’t entirely necessary either, as my service provider has zero problems in this aspect (it can slow you down some, more so, unless you pick VPN servers that are close to the server you play on, but you certainly won’t have that many connection issues in comparison to other alternatives.) If there are also issues with censorship in the country you live in, this is probably a useful alternative, not just to resolve connection problems, I mean I don’t live in the US obviously (your results may vary.)
NOTE: Not every rule may apply here, and in some instances you could take a latency hit of about 5-30ms even, although it won’t really affect download speeds. It has actually increased download speeds for some things on remote servers overseas, such as drivers or source code for a particular framework (where I faced tons of traffic bottlenecks and it slowed to 20Kbps.) You might have to change your settings a lot to get actual results, and in some areas this may slow you down.
The ONLY real speed increase comes from having your own stub resolver or caching-only DNS name server. Everything else is temporary in nature, and requires constant changes to maintain, but there is a way around this problem you’re talking about, which is a connection issue, that’s all I’m saying. It’s EXTREMELY common, that is the only reason why I replied to this with a detailed response. You could solve this problem if you wanted on your own. In actuality, it is NOT Blizzard’s fault, but it would require some creative problem solving on your part.
1 Like
This current problem is not due to DNS.
2 Likes
same here cannot connect takes a while
Same issue. I eventually get in, but there is a long delay. Moonguard, playing in Kentucky.
Another voice in the wind. Same issue.
I think it’s great that there are people with the technical ability to surmount the problem, but the fact of the matter is, it started after server maintenance. I’ve been playing since beta and Retrieving Realms has never been like this.
Add me to the list. Started with yesterday’s maintenance. This is a problem on Blizzard’s end. Whatever you changed has borked the login process. Takes forever anytime you have to access the realm list…
1 Like
I am getting very slow log in’s as above posters this morning. I do get in but it takes a minute ish sitting on the retrieving realm list.
Now I cannot connect, shows the retrieving realm list and then after 30-60 seconds I get the reconnect button. Also taking long time to log into game, hangs at like 60%.
From Philippines west coast server Nesingwary
I read this yesterday (it would have a HUGE impact on most of South-East Asia)…
“Vietnam internet cable broken third time in a year” (VnExpress International)
That was over a month ago and affect Vietnam to Hong Kong. This started yesterday after maintenance and after i’m in my latency is normal.
OpenDNS, Google, etc, are pretty slow. Cloudflare is fast. EVEN if you changed your DNS, it only allows you to resolve the address to a particular endpoint (on a network.) It does not change whether or not your network stack is able to open a socket, and get a response on the other end, or what may happen in between, how the PHY layer is going to respond to any of these problems which present difficulties at one endpoint or another (lots of dropped packets, in some situations where there is enough bandwidth, an entire change in the encoding scheme and splitting and combining of channels, precarious results either way you look at it, other than packet losses or delays due to routing issues, which are beyond the control of the game developers I’m sure.) A VPN can temporarily solve these issues, but if you have other issues that dramatically increase packet loss, on one route or hop in between you and the game server, there’s little that will change on your end, even with a VPN, until the physical issues (which rely on the principles of signal integrity,) are solved. It has very little to do with Blizzard. EVEN if you had login issues, it couldn’t necessarily be related to over-saturation of one channel or another, on a digital packet-switched network / transport medium, you have to look at the fact that they can’t influence anything going on between you and the server (that’s an issue directly related to the service provider, your ISP I mean, but also the content delivery networks / CDNs external to Blizzard’s servers.)
1 Like
Latency is just the ping / response time between the client / server. It doesn’t mean you have higher bandwidth (if you want low latency you need fast upload speeds. Very few ISPs give you BOTH fast upload and download speeds. It’s a little bit deceiving.) You could actually have a low-end internet package and still have performance that matches something 3-5x more expensive. You could have really low bandwidth, and a low sample rate, and still be able to log into the game. You could have really really high bandwidth, but high packet loss, and you would be stonewalled at the login screen every time, just due to a few bad hops on one route between you and the game server itself. If you were able to open a socket 25-100KM from the server itself, I’m sure you wouldn’t see as much packet loss due to the fact that your routing options are more flexible in this regard (hence my explanation of the temporary advantage provided by some VPNs + DNS stub resolvers / caching-only name servers, etc. The negative aspect is that it may hit your latency.) In a digital system if you lose a lot of packets like this, NOTHING works (everything has to be resent, or you have to start all over again.)
Same here…Missouri.
Having the same issue, changing DNS and deleting WTF and rest of folders did absolutely nothing.
Same here, Proudmoore, East
Howdy!
There was a known issue this morning with excessive waiting times on the Retrieving Realm list step, which should now be resolved. If you continue to run into any connection issues make sure to review our troubleshooting Here, then open a new thread if problems continue. Since this issue is now resolved I’m going to close this thread, so we can monitor new reports.
Thank you.
3 Likes