FusionAuth
    • Home
    • Categories
    • Recent
    • Popular
    • Pricing
    • Contact us
    • Docs
    • Login

    Performance issues after upgrade

    Scheduled Pinned Locked Moved Unsolved
    Q&A
    2
    3
    877
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • P
      paul.fink
      last edited by

      Hi,

      We upgraded from 1.40.2 to 1.41.3 and are seeing massive performance issues. Everything is very slow. Signing in, clicking on any pages in the admin side takes several seconds. Doing an api call to get a user takes over 6 seconds.

      The only error in the logs I am seeing is this one:

      2022-11-25T07:54:57.944-07:00	java.io.IOException: Broken pipe
      2022-11-25T07:54:57.944-07:00	at java.base/sun.nio.ch.FileDispatcherImpl.writev0(Native Method)
      2022-11-25T07:54:57.944-07:00	at java.base/sun.nio.ch.SocketDispatcher.writev(SocketDispatcher.java:66)
      2022-11-25T07:54:57.944-07:00	at java.base/sun.nio.ch.IOUtil.write(IOUtil.java:217)
      2022-11-25T07:54:57.944-07:00	at java.base/sun.nio.ch.IOUtil.write(IOUtil.java:153)
      2022-11-25T07:54:57.944-07:00	at java.base/sun.nio.ch.SocketChannelImpl.write(SocketChannelImpl.java:563)
      2022-11-25T07:54:57.944-07:00	at java.base/java.nio.channels.SocketChannel.write(SocketChannel.java:642)
      2022-11-25T07:54:57.944-07:00	at io.fusionauth.http.server.HTTPServerThread.write(HTTPServerThread.java:298)
      2022-11-25T07:54:57.944-07:00	at io.fusionauth.http.server.HTTPServerThread.run(HTTPServerThread.java:158)
      

      We run this in a docker container in aws. I have tried running more instances of it but that did not change anything. CPU load will get high if anyone tries to do more than one thing with the system. Memory stays the same. Any ideas on what we can do or what changed?

      P 1 Reply Last reply Reply Quote 1
      • P
        paul.fink @paul.fink
        last edited by

        I would like to add that we have a production instance that is running with less resources with no issues. The production instance has way less users, groups, registrations, applications, etc. So the data seems to have something to do with it.

        joshuaJ 1 Reply Last reply Reply Quote 0
        • joshuaJ
          joshua @paul.fink
          last edited by

          @paul-fink marking this thread as being addressed out of this forum band.

          -Josh

          1 Reply Last reply Reply Quote 0
          • First post
            Last post