BUG - Degraded server performance srv2 EN

  • We are having trouble playing on this server after the merge. :thumbdown:


    It was ~fine in the beginning, but over time, it has become slower and slower. It now takes around 5 seconds for the loader to accept a cultural treaty or open a building.

    In most cases, it takes 5-10 seconds to switch between towns, and sometimes more than 20 seconds to scroll through islands.

    Please look into this before it becomes completely unplayable.



    • Name: Perseus
    • Speed: 1
    • Time of departure : 09.09.2024
    • Special Server: No
  • Minotaurus is also affected. I believe it's because of large populations on both servers and it may be worse over time.

    However, Persephone is fine, considering it's the least favorite server atm.

    segimann2.png


    Tusen takk til Infinity for dette stilige bildet!

  • At the moment, the servers are busy collecting tons of data from the accounts that are moving there.


    This will improve over time as the number of accounts to be moved decreases.


    Additionally, the activity is very high.

    We have already seen this with EN_401 and the activity there.


    The issue is being monitored continuously.

    lagu-ika_admin.jpg

    Sig und Ava made by Butterfly I

  • Totally agree with this topic, with every click I do I can count to 7 and then it loads. Really depends on the time of day you are playing.
    Is it possible to give some servers more capacity? Because it really is unplayable at this moment.

  • By any chance is are there memory leaks within the server code base that have normally been handled with a daily reboot? All the additional activity is likely to result in the need for more than one daily reboot if such leaks are being handled as just part of a daily cycle.


    Having a database that needs to rebalance its data structures during quiet times will also cause the same problems if there is no available quiet time.

  • The downfall of ikariam :closedthread:


    It is probably not something they can fix by adding more hardware power since they would probably do it,
    the problem is in poor implementation that that has now come to light.

    We have been warned with ambush event :)



    Is this issue going to be tackled or just continuously monitored if it continues after the merges?


    If not please let us know so we can quit our efforts to play it.

    Edited 2 times, last by Vector ().

  • Perseus is now stuck in something of a death loop at the moment - it takes so long to reply to a request that the player has already made more requests that it then has to process - repeat until it takes so long that whatever timeout is coded is reached.

  • What are you saying! Are you trying to blame something that doesn't exist?


    Please correct the mistake before people leave the game for good...

    7af75d9f45e0d7e43eedeaeddecdc978.jpg

  • It seems that yesterday we were impacted by the merger of the TR, BG, US servers.


    So all the different instances are being deployed on the same physical equipment (maybe as VMs) so the heavy workload caused by a merger goes and impacts some/all of the already merged servers trying to get their worlds back up and running.