Uploaded image for project: 'Dev - Nexus Repo'
  1. Dev - Nexus Repo
  2. NEXUS-10135

NX3 killed during 640Gbyte repository migration - OrientDB consuming all available memory

    XMLWordPrintable

    Details

    • Type: Bug
    • Status: Closed
    • Priority: Major
    • Resolution: Duplicate
    • Affects Version/s: 3.0.0, 2.13
    • Fix Version/s: None
    • Component/s: Migration
    • Labels:
      None
    • Environment:
      AWS Performance testing environment, repository migration of 640Gbyte repository.

      Description

      While executing 640 GBytes repository migration test, NX3 process was killed crashed after about 14 hours.

      1. Largest component in db is asset.pcl
, size 9.907Gbyte.

      2. Last log entry in nexus.log is

      [gc][young][50026][91606] duration [1.7s], collections [1]/[3.5s], total [1.7s]/[23.1m], memory [872.1mb]->[877.7mb]/[1.1gb], all_pools

      {[young] [345.2mb]->[3.3mb]/[381mb]} {[survivor] [6.4mb]->[4mb]/[9.5mb]}

      {[old] [520.3mb]->[521.3mb].

      An investigation is required to assess if it is a memory leak or an out of memory condition on the JNI interface.

      The proposed approach is to re-run the test with GC verbose and Jni verbose options.

        Attachments

        1. nohup.out
          601 kB
        2. pmap.out10
          58 kB
        3. report-404-04251461609072.pdf
          201 kB

          Issue Links

            Activity

              People

              Assignee:
              aavritzer Alberto Avritzer
              Reporter:
              aavritzer Alberto Avritzer
              Last Updated By:
              Peter Lynch Peter Lynch
              Votes:
              0 Vote for this issue
              Watchers:
              2 Start watching this issue

                Dates

                Created:
                Updated:
                Resolved:
                Date of First Response:

                  tigCommentSecurity.panel-title