Server Stability must be Fixed


  • This is more feedback rather than an idea, but it is something that must be addressed!

    I've been playing on a couple public servers, and I've got to say that the stability of these servers simply needs to be fixed. The servers need to be reset on a daily basis, sometimes twice a day or even more, and to me this is simply unacceptable. If there is somebody running that server who is available to do the reset, it isn't terrible (still unacceptable) but playable.

    I get that this is an alpha release. I get that there will be bugs and that potentially the game could be unplayable for awhile. I also get that the developers of this game are trying hard to improve this game and that some things can slide where priorities happen elsewhere and sometimes bugs are going to be hard to find as well.

    All of this changes though if as is the plan that the game will be released to a much wider audience with the Steam Release and that this game is going to be called "Beta". It doesn't mean that it needs to be perfect and that there can still be some occasional issues in terms of things not quite working right.

    Beta doesn't mean though that the game is unplayable.... especially a public beta. While not feature complete, the levels of professionalism are going to need to be much higher and the quality of simply using the software (not necessarily gameplay elements or how it interacts.... but simply UI issues and being able to play the game at all in any form) has to work. I can't emphasize this enough..... it must work. When you click the button to cause it to run, it will work and when running a server players can't get kicked out for light or everyday reasons that are unexplainable.

    If this was just a single server that was having configuration issues or had connection problems.... that could be addressed by the person deploying the server. That isn't the case with this gripe here.

    My fear, because I really do like this game and because of what I really think the developers of what has been done so far with Eco need kudos for their hard work, is that when the Steam release happens that this game will get so many negative reviews and so many people complaining about the unplayable nature of the game so far that nobody will buy the game and that it will fade away as a failed effort. I don't want to see this game crash and burn. I could name some specific examples of games on Steam where this has happened... but I'll avoid advertising those negative examples and hope that Eco doesn't face that fate.

    And no, putting the server reset on a chron job is not sufficient. That is a cheap hack and not what is needed. Real bugs need to be addressed and when you run the server it shouldn't bug out for people playing that are hesitant about playing and don't really know what should be happening or not.


  • 13
    Posts
    1057
    Views
    Log in to reply


  • tl;dr;

    Without reading the full post, I am assuming what you're saying by reading a few paragraphs,

    However, please note that the devs are currently working on catching the exceptions for the servers not too crash but to silently catch these and handle them.



  • However, please note that the devs are currently working on catching the exceptions for the servers not too crash but to silently catch these and handle them.

    That is a lousy way to deal with bugs.... and not what I'm talking about. I've done software development (still am) and catching exceptions isn't going to handle all of the issues I'm seeing.

    Stuff like how players are working and then suddenly not able to place a block while other players can do that, or when performing a trade it freezes up are likely not going to be "caught" with an exception handler. Those are hardcore logic bugs that aren't going to be solved until the issues which create those conditions are dealt with properly with sound algorithms and logic.

    There is a time and place to set up exception handlers and circumstances you shouldn't.

    If this was continuing to be an alpha release for the next several months and not going to steam, I wouldn't even bring this kind of thing up. Work through those issues and make the game more stable with its code base while pulling out spaghetti code and cleaning stuff up that was an early programming experiment that is no longer needed. That kind of debugging isn't glamourous and doesn't show up much on bullet points when doing a new release, but is important overall for any software development project.

    And no: it wasn't too long. If you think that is a wall of text, you are simply too young and inexperienced.



  • You need to remember that, except for one server, Eco devs have no control over all these servers. These are all owned by individuals. My servers are all scheduled to restart daily, others restart as needed and others only restart if they crash - this is up to the individual Admin to set up a auto restart to do this.

    As Eco gets more stable it will not need this much. In the meantime the Admin can set up a scheduled task to auto restart. This is an ea game under development and because of that it will have stability issues. There is no way to be 100% stable when you are changing and adding to and working the code. Steam users will be told that if they want a stable game they should wait for final release.

    Pam
    Community Moderator
    Discord: Pam#5439
    reddit moderator: rentechd
    https://discord.gg/Mg4WJe9



  • I am not talking about the Eco devs having control of the servers.

    I am talking about fixing the server code itself

    Eco needs to get far more stable so those operating the servers don't have to constantly be harassed by their players and the community of folks on a server to be resetting the server on a daily basis. That even includes your server Pam. It is one of the reasons I left.

    No, an autostart is not sufficient. LIke I said, a daily chron job of restarting the server is simply a cheap hack and avoiding software quality assurance and fixing bugs that need to be fixed. It is professionalism that will be expected for those on Steam.

    If this is supposed to be normal, then don't release on Steam. Hold back on the Beta release until these bugs get fixed.

    If it continues in this fashion, Eco will last a whole two weeks maximum on Steam before it will be dropped like a stinky pile of fertilizer with so many negative reviews that the devs will not be able to sell any more copies on Steam. It will be a failed effort to even try to use Steam as a source of revenue.

    Go ahead and ignore my advise and criticism. You don't owe me anything at all, but I'm simply offering this warning and suggesting the Steam release won't go very well unless stuff like this is fixed and fixed properly. I wish the devs well, but this is the kind of thing they should ignore at their own peril. It can be a permanent black eye to the whole project if that release doesn't go well.



  • HI Korihor,

    Although I partially agree with your previous statement, I have to disagree,

    While solving bugs is not catching exceptions, MOST of the crashes I see on many servers are easily captured via a catch, and atleast until the devs fix this issue, the players don't get bored of the crashing and restarting.

    So, I know that fixing bugs is not catching the exceptions. I am a software engineer, and have been for 11 yrs, but there are cases like this where the user base are getting pisse doff by the random crashes. These can be avoided and later handled.

    Anyways, the bug you mentioned, has never happened to me before, unless if the connection between the client and the server was ended. And after a few seconds you get the Inactivity message.

    Thanks.



  • The only reason you use a "Try.... catch" type of expression in your software is to deal with things you know are going to happen and for it to be deliberate and intentional. It usually has something to do with either thread management or perhaps some specific bit of code that is deeply nested within a whole bunch of subroutines you want to pop out immediately.

    I'm not saying it is something to completely avoid using like a GOTO type statement as it has its uses in programming, but to throw such programming structures around a block of code you know to be buggy is just simply being lazy and produces unpredictable results. That isn't solving bugs, it is whitewashing them and hoping they go away.

    And no, it doesn't necessarily end the connection between the client and the server in the situations I've seen. Perhaps I should post about it as a specific bug report, but it is a common enough occurrence that other players on multiple servers I've been on have mentioned it. It isn't the same thing and a server spontaneously shutting down or freezing... although I've seen that happen too. I've had the servers continue to operate on their merry way and have even multiple people log in and out... yet basic things within the game won't happen. Sometimes it ends the connection, sometimes it doesn't.

    This appears to be a whole class of bugs and not a single one.



  • From my experience, when software / features are in early stages, we always add the try catch method, and we log everything into a database, so we can monitor any expceptions the users are experiencing.

    This is the best way to go about alpha's. For you to say otherwise, it shows how much lack of experience you have in the field. This is alpha, anything for them to catch bugs and help them fix them, kinda goes.

    We are here to help them test the game, so if you notice a bug, report it!

    Unless someone reports it, devs might not experience it and therefore might not fix it.



  • When doing an Alpha? Perhaps. I'm not talking about an untested alpha pre-release candidate here. I'm talking about what needs to happen when it goes into Beta or more importantly because those on Steam really don't care: it will be a major public release.

    You say it shows how much lack of experience in the field I have? I guess I don't have much in the way of game experience where stuff like this can slip and perhaps people at a place like Electronic Arts doesn't give a damn about its customers. Instead, I've had to program stuff for millionaires that simply don't give a... heck... about your job if you screw up with stuff like this. You simply get it right the first time when they have to use it. You do quality assurance testing on multiple levels and if needed you even do proof of the algorithm.

    Should I report some of these bugs? Perhaps. I don't have access to server logs or save files since these aren't my servers I'm using. And BTW, consider THIS to be a bug report as well. If the devs don't read these forums, they should shut this forum down instead.

    I'm not expecting Eco to have the same quality of performance as perhaps a piece of medical equipment or something in aerospace, but I am expecting something of perhaps a bit more quality than what I'm seeing here. If you want to keep this in Alpha and keep the community smallish who can work through bugs on this level, then go ahead and keep the existing practices and standards that I'm seeing. It won't be good for the game or Strange Loop Games if it continues on like this for the Steam release though. If you don't want to pay for quality assurance testing, then you get what you pay for too.

    You are expecting a whole lot from unpaid volunteers to be demanding they do anything. Perhaps I might submit a formal bug report after digging through a bunch of pages to find out where the devs might be wanting to see that bug report. The harder it is to do that, the less likely I will be to make such a report.



  • Hey King, we're working hard to make this the case. In fact today is 6.4 release and 'feature lockoff', so from now till Feb 6 steam launch we're focusing only on bugs and stability. I agree its super important to be stable on Steam, thats why we've waited this long to get it on the platform.



  • And yes reporting bugs is super helpful, please do: https://github.com/StrangeLoopGames/EcoIssues/issues

    We've had a lot of reports from the community that got fixed in 6.4, really appreciate people's help there.



  • Thanks for the reply, John. I had wanted to respond to Korihor's original post, but spam filters were blocking me. Anyway:

    Server stability really does need to be a top priority. I am the admin of the server Korihor is referring to, and there have been all sorts of "soft" crashes, such as crafting being halted, certain users being constantly kicked with an "Inactivity" message, or players not being able to place blocks. To the extent that these types of game-breaking issues can be prioritized, they must be.

    I think Korihor's heart is in the right place. Some of us are just more... pessimistic about Steam users' standards for an EA title with a $30 price tag. Sure, EA games have been around awhile and people will expect bugs. But if there are game-breaking bugs, they will complain, and possibly give negative reviews. None of us want to see that happen.

    Some of the server issues I've seen have been hard to reproduce, and are complicated by the fact that I've sometimes used admin tools for playability reasons (used Dev Tool to destroy a bugged crafting station, for example. Could this lead to a memory leak? I don't know.)

    I'm eager to help out with bug-testing on a "clean" server in 6.4, though!



  • I'm afraid I'll have to agree on the Early Access thing, major bugs will crash the reviews. It's an inevitable fact. Many people are very quick to judge and won't tolerate any game-breaking bugs more than once or twice (if any at all) before leaving really bad reviews.


13
Posts
1057
Views
Log in to reply

Internal error.

Oops! Looks like something went wrong!