Uploaded image for project: 'WildFly'
  1. WildFly
  2. WFLY-12167

Memory leak in metrics in standalone-ha configuration

XMLWordPrintable

    • Hide

      1. start pristine WildFly 16 with -c=standalone-ha.xml
      2. send it a bit more than 10.000 requests to /metrics (equivalent to about 1 day of runtime while being scraped by 2 prometheus instances in default 15s intervals)
      3. observe increase heap memory usage and ultimately OOM

      Show
      1. start pristine WildFly 16 with -c=standalone-ha.xml 2. send it a bit more than 10.000 requests to /metrics (equivalent to about 1 day of runtime while being scraped by 2 prometheus instances in default 15s intervals) 3. observe increase heap memory usage and ultimately OOM

      When started in standalone HA configuration every request to the recently added metrics endpoint (<management-if>:9990/metrics) lead to an increase in memory consumption until the JVM is slowed down significantly by GC to a point where even the requests to /health fail within a reasonable timeout (2s) and untlimately lead to OOM.
      The same issue does not occur when WildFly is started in the default standalone configuration (non HA).
      I can provide a (compressed) heap dump if required.

        1. Screenshot 2019-06-06 at 11.07.00.png
          Screenshot 2019-06-06 at 11.07.00.png
          374 kB

              ropalka Richard Opalka
              bsx303 Bernd Stolle (Inactive)
              Votes:
              0 Vote for this issue
              Watchers:
              5 Start watching this issue

                Created:
                Updated:
                Resolved: