narukthul Posted January 17, 2019 Share Posted January 17, 2019 (edited) For the last month or so now HH servers have been so terrible that it takes 5 minutes just to collect from the girls because the server goes down ever 5 girls. Why have the HH servers gotten so bad? They were never that great but weren't nearly this bad. If this keeps up a lot of people are going to shelf the game. Edited January 17, 2019 by narukthul Link to comment Share on other sites More sharing options...
Cantrix Posted January 17, 2019 Share Posted January 17, 2019 Dunno, but it's terrible! I can hardly nake 3 boss fights within 2 minutes. Though i should mention that i started to experience hjgh latency and unresponsivenes ever since the last patch hit a few days ago, not a whole month before that. Link to comment Share on other sites More sharing options...
narukthul Posted January 17, 2019 Author Share Posted January 17, 2019 idk, I could be exaggerating about a month. feels like it with the slow servers though lol Link to comment Share on other sites More sharing options...
narukthul Posted January 17, 2019 Author Share Posted January 17, 2019 503 Service Unavailable No server is available to handle this request. Blah... 1 Link to comment Share on other sites More sharing options...
Urist Posted January 17, 2019 Share Posted January 17, 2019 They're pretty bad, but they're always especially bad at league reset time, at this very moment. Link to comment Share on other sites More sharing options...
narukthul Posted January 17, 2019 Author Share Posted January 17, 2019 It's more than just the league resets though. been bad constantly for at least a week 1 Link to comment Share on other sites More sharing options...
Zorba the Geek Posted January 19, 2019 Share Posted January 19, 2019 This is a huge gripe of mine. I've sacrificed arena and ToF battles because of this and I dread having to collect money from my harem because of the time it takes to collect it. As for performances against bosses... the less said about that the better otherwise I might use a bad word! Suffice to say, it has cost me the equivalent of two rounds (40 performances) trying for Bunny's Mum last revival. This might well have cost me any chance to recruit her in my Bryan account which played up the most. For me the most troublesome time of the day is variable, but mornings (my time) are generally the worst. I can't specify which action is the worst because each one has its turn and can be anything from collecting money from my harem to selecting items from the market (and using or selling them), performances against bosses, logging out and logging back in, selecting Pachinko (although this has much improved) and getting the current quest to move to the next scene. In other words, just about every aspect of the game. It must be almost as perplexing to HH developers as it is for us, the players, otherwise it would've been fixed by now. I suspect it is the price of popularity and an extreme level of data traffic as a consequence. Link to comment Share on other sites More sharing options...
saberbholt Posted January 19, 2019 Share Posted January 19, 2019 (edited) 3 hours ago, Zorba the Geek said: This is a huge gripe of mine. I've sacrificed arena and ToF battles because of this and I dread having to collect money from my harem because of the time it takes to collect it. As for performances against bosses... the less said about that the better otherwise I might use a bad word! Suffice to say, it has cost me the equivalent of two rounds (40 performances) trying for Bunny's Mum last revival. This might well have cost me any chance to recruit her in my Bryan account which played up the most. For me the most troublesome time of the day is variable, but mornings (my time) are generally the worst. I can't specify which action is the worst because each one has its turn and can be anything from collecting money from my harem to selecting items from the market (and using or selling them), performances against bosses, logging out and logging back in, selecting Pachinko (although this has much improved) and getting the current quest to move to the next scene. In other words, just about every aspect of the game. It must be almost as perplexing to HH developers as it is for us, the players, otherwise it would've been fixed by now. I suspect it is the price of popularity and an extreme level of data traffic as a consequence. I do not know where you see it must be strange for the developer, how much it can be expensive but strange First they know how much the data packet sent by a user and the one that answers that user occupies from the cluster of servers, then users connected at the same time x size of the user packages = connection download of the cluster; changing by packet size sent by the server = connection upload cluster. If a symmetric fiber connection of 300 Mb is not enough, you put more similar lines attached to a load balancing server to manage the upload and download available. With that if it is by bottleneck in the line it is solved. Step 2. Size of the database and management of requests. The database is divided into blocks to be loaded into the ram of each cluster machine; for example, of the users who have connected in the last month in alphabetical order the first 10,000 records to the first PC of the cluster, the next 10,000 to the second, etc. A machine is created from the cluster to verify the user's name and the request is sent to the pc that has that record in the database, that pc manages the operation and sends the response to the "verify user" PC and the latter send the request for sending to the load balancing server to use one of the upload lines that have available bandwidth. If it is a user who is not connected in a month, the file server is searched where the database is and after completing the operation the result is returned to the "server to verify user" to do what in the previous case. If the number of requests is greater than what the "verification server" can manage, more machines are added to that node with load balancing to process the petrifications and redidections. Step 3, everything is connected to the maximum possible speed, use optical cards with optical fiber to avoid creating bottlenecks. And this is how a problem of requests and databases is solved, for which I indicate that it is not usually strangeness, but an infrastructure and monetary problem and it is a scalar problem, to more users use the service, more connections and operations in the server, for which you need a server created with a cluster that allows to scale at the same time as the demand grows is a bit more complex diagram, here I put the basics, keep in mind that for example the server load balancing lines should also be an "isa server" and provide security against attacks DDos, etc; On the website is the forum page, for example. But the general idea is that. Edited January 19, 2019 by saberbholt Link to comment Share on other sites More sharing options...
jelom Posted January 20, 2019 Share Posted January 20, 2019 @saberbholt If HH1 error were caused for a problem in the servers or the servers' line then it wouldll affect all players equally, and that is not the case. There are some players who almost never have a problem (it is my case, fortunately) and there are some players who can hardly play because of it. To be fair, I must say that lately there is more lag sometimes, so perhaps servers are a bit overloaded at that times. The 503 error at the league start (Thursdays at 5:00 CET) it is a different problem and it is affecting all players, so it is sure in the server side, probably overload because many players are hammering the server doing their 15 challenges. Link to comment Share on other sites More sharing options...
saberbholt Posted January 20, 2019 Share Posted January 20, 2019 (edited) 1 hour ago, jelom said: @saberbholt If HH1 error were caused for a problem in the servers or the servers' line then it wouldll affect all players equally, and that is not the case. There are some players who almost never have a problem (it is my case, fortunately) and there are some players who can hardly play because of it. To be fair, I must say that lately there is more lag sometimes, so perhaps servers are a bit overloaded at that times. The 503 error at the league start (Thursdays at 5:00 CET) it is a different problem and it is affecting all players, so it is sure in the server side, probably overload because many players are hammering the server doing their 15 challenges. The error HH1 erro r is caused by packet loss in the connection, and many users do not happen because they use different nodes in the connection by their isp, because they use different browsers that manage differently the packages and the times before giving the packets for lost, because not all the mindo uses the same DNS to resolve the connection before being routed to the node of your ISP, for how you have configured the protocols of the network card (or what protocols you do not have , to less protocols less size of packages and check), or by how it manages the connections and latencies of hardware the operative systems, or conected with wifi or cable, for all that reason there are different results in different users, but ... All those packages are lost in a large part due to a delay in the response of the server. the best way to see if there is a lot of traffic and the server is saturated by connections is to simply launch a Ping to the server's ip, in my case I never lose a package against the servers of google.com or microsoft.com, against hentaiheroes.com I can not say the same, I do not fall more than 10% as maximum, but they get to lose in times of heavy traffic. you will see that the 37 ms that are now at 10:53 GM +1 do not look at all at 5 o'clock in the morning the new event day for example or the last league coming day, was the error HH1 more than one time and was make a ping obvius lost 80% of the packets, when the some machine and the some cable conection have diferents results with same conditions in one side, the problem is always in the other side (excep problems with nodes of your ISP) The other way is at most have other problem in server side with the time until the data base close automatic a conection for consulting sql dates or multiple conections, that can be too a problem. The 503 error is more a web server down but that referies at apache server for the web or the host site and in any chases server down Edited January 20, 2019 by saberbholt Link to comment Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now