Hi @Caterpepi,
I have tried all of the above, as well as everything else I can think of, to no avail. As I understand it, this is an issue for many people (every 5th post of this form basically). I think it unreasonable to expect a large number of customers to try all this troubleshooting to get buggy code to work. Please escalate to the technical team to fix it on hearthstone side.
Thanks
This morning, for the first time since this started months back, I cannot seem to establish a connection at all. I assume this is because things are slow server side due to maintenance. It also points to the fact that it is server side issue.
2020-04-28 13:08:33.008: [BattleNet] [SslSocket] Exception while trying to call EndRead. System.IO.IOException: EndRead failure —> System.Net.Sockets.SocketException: An existing connection was forcibly closed by the remote host.
at System.Net.Sockets.Socket+SocketAsyncResult.CheckIfThrowDelayedException () [0x00000] in :0
at System.Net.Sockets.Socket.EndReceive (IAsyncResult asyncResult, System.Net.Sockets.SocketError& errorCode) [0x00000] in :0
at System.Net.Sockets.Socket.EndReceive (IAsyncResult result) [0x00000] in :0
at System.Net.Sockets.NetworkStream.EndRead (IAsyncResult ar) [0x00000] in :0
— End of inner exception stack trace —
at System.Net.Sockets.NetworkStream.EndRead (IAsyncResult ar) [0x00000] in :0
at Mono.Security.Protocol.Tls.SslStreamBase.InternalReadCallback (IAsyncResult result) [0x00000] in :0
at bgs.SslSocket.ReadCallback(IAsyncResult ar)
Aye, it is THEIR battle.net server forcibly destroying our SSL connections (as per the logs). Support is actively ignoring all of us.
Chuchu,
Moved you to your own thread so we can work with you individually. Those logs don’t have anything to do with the server itself. One of your sockets got closed down due to a timeout while trying to contact our servers. That’s normal behavior for a connectivity socket when you can’t establish a connection. It closes back down.
Given that, the key here is going to be figuring out why you’re timing out. Given where you live, I’d wager it has something to do with an ISP between you and our servers, since you have a long way to go to get to our servers and a lot of ISPs are having issues right now with the coronavirus lockdowns. The easiest way to test this is to get a connection test.
If you’re still having this problem, please create a WinMTR test which catches a disconnection. You want to restart the test every 10-15 minutes until you catch one of these problems. Once you’ve caught one red handed, run the test for about 5 more minutes. This will let us look for problems between your house and our servers which may cause this.
If you have problems with posting it due to a link error, go ahead and copy paste this into your next post, and replace “WinMTR goes here” with your test.
~~~~
WinMTR goes here
~~~~Thanks,
I already caught it in the logs though?
[BattleNet] [SslSocket] Exception while trying to call EndRead. System.IO.IOException: EndRead failure —> System.Net.Sockets.SocketException: An existing connection was forcibly closed by the remote host
I’ll do the MTR thing though to see if we can get more detail.
Thanks for getting back!
PS. it is standard to SSL to drop but the reconnect should be instant and seamless. you cannot expect an entire restart each time a connection jitters. No other app of game does this, for a good reason. Please, this is a widespread issue and only recent. They need to roll back whatever change it is or increase timeout to something sensible. I can play an entire day of WoW even with this battle.net issue, and I always finish my games in HS, so why make us restart the game if it has already reconnected?
Basic test:
Ping
1ms
Download
50.72Mbps
Upload
51.04Mbps
MTR:
|Host|%|Sent|Recv|Best|Avrg|Wrst|Last|
|---|---|---|---|---|---|---|---|
|router|0|490|490|0|2|67|0|
|fc-empire-clients.faircom.net|0|490|490|0|2|138|1|
|fc-empire-core.faircom.net|0|490|490|0|2|78|0|
|131w.residential.faircom.co.za|0|490|490|1|2|86|1|
|gwf-tdc3-core1.faircom.net|2|468|462|1|3|106|1|
|gwf-tdc3-edge1.faircom.net|1|482|480|1|2|110|4|
|195.66.226.234|2|470|465|143|149|342|183|
|ae1-br01-eqld5.as57976.net|2|458|450|149|153|272|151|
|et-0-0-2-br02-eqam1.as57976.net|3|443|431|148|155|274|149|
|137.221.78.85|1|486|485|149|153|314|149|
|137.221.66.43|1|482|480|149|151|225|149|
|185.60.112.157|**1**|478|475|149|150|216|150|Heya chuhu,
Thanks for providing this test. This shows what I was concerned about. It looks like as you go through the faircom backbone provider you start dropping around 2% of data packets, which is why your reconnect isn’t working when this happens. If you check the test, you’ll see that after this node, you consistently are missing packets in the test, indicating that they’re just never making it to our servers. The game client will try to reconnect a couple times when you have a timeout, but if you drop too many packets in a row, or fail to re establish the connection with our next few attempts, we drop you.
Out of curiosity, have you tried a VPN to work around this? I’d be interested to know if it fixes things. There are a number of free ones out there that you could test out to see if it helps. Note that because of your location, you may need to try a few or see if there are any that gamers in your area like to use to improve connectivity. Often with a backbone issue like this, using a VPN will take you on a detour around the overloaded part of the network. Since the COVID-19 quarantines began, we’ve gotten a number of reports from players around the world that things like this are happening more frequently than normal. Most folks are having to resort to VPNs to work around the parts of the internet are strained. Note that you’d take a different path to our servers than others, so it’s even possible for you to have no other symptoms if you happen to take different routes to other services you use regularly.
Thanks Drakuloth,
I’m on the NordVPN Amsterdam VPN server now and made it through 10 games without crashing.
Still though, there are many dropped packets to the server. This is a normal thing. The PROBLEM HERE IS the functional bug in HS that, after the server has long since reconnected in the background and after the game has finish several minutes later, HS then boots us for the dropped connection minutes before that is long since reconnected. Please, this is what needs fixing, we understand the general connectivity is bad and out of Blizzards control, but that wouldn’t be an issue if you just fix that functional bug. Please at least pass this thread forward to technical, even if I’m not involved? It would save a lot of people unnecessary work-arounds at a time where, understandably, support is very thin.
Here is my MTR from the 10 games:
|--------------------------------------------------------------------------------------|
| WinMTR statistics |
| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |
|--------------------------------------------|------|------|------|------|------|------|
| 10.8.0.1 - 3 | 3057 | 2993 | 160 | 168 | 1799 | 165 |
| zdfgdfgdfzgdf - 3 | 3061 | 2998 | 160 | 168 | 1572 | 164 |
| zdgdfgzdfgzfdgf - 2 | 3136 | 3092 | 160 | 172 | 2074 | 164 |
| adfgrdfgdfg - 2 | 3129 | 3083 | 161 | 176 | 2067 | 166 |
| zdfgdfgdzg - 4 | 2931 | 2839 | 162 | 193 | 4691 | 360 |
| 137.221.78.87 - 2 | 3156 | 3119 | 0 | 184 | 2140 | 170 |
| 185.60.112.157 - 1 | 3180 | 3149 | 167 | 183 | 2092 | 171 |
|________________|||||||
Chuchu,
While I can respect that you want to give feedback, this is the technical support forum, and we here are not developers. My job is to get y’all playing and get you any technically relevant information I can. We tech blues are CS agents, and don’t work directly with Developers.
If you want to submit a bug report for reconnect code or submit a suggestion, you’ll want to go to the correct forums for those, as we don’t go into the meetings where they deliver this feedback. We’ve got a separate team that combs forums for game feedback. Glad you’re able to work around it with the VPN, though!
Hi Drakuloth,
I can appreciate that. I’ve tried all the channels, and have an ongoing ticket, and gotten nowhere so far, unfortunately. It is clearly and issue that there is a fundamental disconnect between those on support and those who can fix the issues.
VPN trick only helps about 30%. Still frequent disconnects.
UPDATE:
I finally got some clarity from a technical person via 9th web ticket. Sent a 10th asking for more detail and why the auth connection is the only one that fails.
Hello Adventurer!
Gamemaster Khaleyd at your service - I’m with the dedicated tech team here at Blizzard. Thanks for getting back in touch and letting us know about your thoughts. It does appear we failed in explaining what actually happens in these cases, my apologies for that - I will try to shed some more light onto the situation.
The particular situation you described occurs when your connection to the game server (that your match occurs on) is still functional while at least part of another connection route that Hearthstone relies on (typicaly towards the radius/auth-server) has already been disrupted. In that case the game will usually still be able to finish your match, but once it has to pull new data from your account (e.g. to update how your stats changed) the missing secondary connection becomes apparent and starts causing trouble. This is also reflected in the error message - your stats get saved, and then a new connection towards all of services the game utilizes is required to progress further. The previous connection route, which has already long failed at that stage, is then indeed terminated to allow a new route to be established - this is basically what you spotted in the log mentioned in your first message, although not very clearly spelled out there.
To be very clear about one aspect here: this is not a bug, but a quite intentional (and necessary) functunality that acts deeply in favor of our players. The system acts a failsave to allow players whose overall connection quality may be somewhat shaky to still play the game, and unfortunately right now (thanks to the ongoing crisis) a great lot of players find themselves into circumstances where said failsave is very much needed (meaning its effect becomes visible a lot more). The system itself is however not new - it has been in the game since day 1, just never was as often needed to come into effect as it has been in the past few weeks.
Hopefully this clears up matters for you a bit. We’d love to be able to tell you that things will be better soon and such failsaves won’t be needed anymore, but realistically speaking it is very likely that while this crisis continues and ISPs keep throttling/de-prioritizing routing towards gaming services such as battle.net that the relatedfsymptoms won’t disappear either.
Kind regards, and I wish you a very nice weekend,
Thomas “Khaleyd” W.
Gamemaster Tech Team,
Blizzard Entertainment
Thank you so much Khaleyd!
How refreshing to get a technical explanation here. I understand much more clearly now.
A few questions if you will allow:
This “secondary connection” - how come it doesn’t fail but the auth connection keeps failing? I don’t doubt what you are saying, I’m just trying to understand why the auth one cannot work similarly to the other connection that never fails. Indeed, no other game I play has this issue, so I’m trying to wrap my head around why HS is different and cannot be made to behave like other games.
Is it not possible to reconnect to auth in the background without restarting the whole of HS? I cannot tell you how much time and frustration this would save many many people with this widespread, growing issue (that, granted, is not Blizzard’s fault). It might be pertinent to adapt to the new realities here.
Many thanks!
Chuchu
Greetings,
I am Gleedoh. I have well considered your request : how are you today ?
Hope you are fine ! 
Thank for your questions
For the question 1, it’s merely a question of a reroute or a secondary connection path. Imagine that the main path is facing an issue, the emergency reroute can help you to retrieve the game and avoid a game lost.
We cannot be more explicit for confidential reasons, but you can be sure that the system is working for you and not against you. 
As for the question 2, we think that this is not possible because, restarting the game means to kill the Hearthstone.exe process and doing that step for sure the game will be lost. The system needs to keep a contact with your Hearthstone.exe.
But, if you have ideas we encourage you to use the forum.
Regards,
Gleedoh, Game Master
Blizzard Entertainment
Good Day,
With all due respect you did not answer either question:
Regards,
Chuchu
Heya!
So the two questions are sort of part of the same issue, in that the “secondary connection” as you put it is part of a different port connection as we understand it. Similar to how login servers and game servers are not all part of the same thing (as to login to the game is a different type of connection to actually connecting to a game server to play against an opponent, for example), different connections also require different levels of authentication.
In terms of automatically reconnecting to both it sounds like one is having trouble with this connection and the other is staying, but in terms of the game forcing one to connect and the other not to this would be a development change (which is something we game masters dont deal with and the developers would need to do manually via feedback on the forum) and possibly not something that can be done as it would require a form of communication through two networks, one with an authentication carrier and one without. At this point this is sort of speculation, as we aren’t programmers on the game, but the assumption would be something along those lines.
Hopefully this helps explain everything, the developers would be the only ones that could truly fill in the gaps and unfortunatley we don’t work with them!
FINAL UPDATE:
OK… so our only hope here is that someone on the development side sees this post… Prayers to RNGesus
Chuchu,
The devs don’t typically go through this forum for suggestions and the like. You’d be better off reposting in General Discussion with a suggestion for a change to reconnect functionality and possibly link back to this thread, as that’s the appropriate process for feedback. Good to see that you got some more questions answered nonetheless, and if you have any other tech related problems that we can help solve from a tech perspective, feel free to reach out!
This crap is happening to me since february, still waiting when some fix will be made. Started out of nowhere so I hope that it will also disappear one day… very frustrating experience. At least there are another games which run just fine.
Indeed… it is not directly Blizzard’s fault, but rather due to the increased global internet traffic in Lockdown… that said, it exposed some poor connection handling in HS code… the biggest issue is that the support on the ground is fundamentally disconnected from anyone who can actually fix issues in the code… this is a crazy oversight by Blizzard; and so we must live with this issue for months to come
Replying to Joza below (since this central thread on the issue has been closed for some reason):
I’ve noticed the lag since the new update only, but the disconnect issue has been months now. We seem to be going in the wrong direction. Clearly playability is not a concern of theirs.
Well it looks like there is not only a connection problem the game lags like hell out of nowhere just when this crap with corona started. I cant even play BG properly with the lags sometimes i cant buy, sell a minion when situation is close it just simply lags out.