Uploaded image for project: 'Karaf'
  1. Karaf
  2. KARAF-7839

Karaf client ssh idleTimeout not working

Attach filesAttach ScreenshotVotersWatch issueWatchersCreate sub-taskLinkCloneUpdate Comment AuthorReplace String in CommentUpdate Comment VisibilityDelete Comments
    XMLWordPrintableJSON

Details

    • Bug
    • Status: Resolved
    • Minor
    • Resolution: Fixed
    • 4.4.3, 4.4.6
    • 4.5.0, 4.4.7
    • karaf
    • None

    Description

      Karaf client idle timeout doesn't work if longer than 60s (default for heart beat).

      This can be reproduced by:

      1. Downloading vanilla karaf
      2. Changing idle timeout in "org.apache.karaf.shell.cfg" to "sshIdleTimeout = 90000"
      3. Start karaf with ./bin/karaf
      4. Start client in separate terminal with ./bin/client

      Suspicion is that the bug was introduced here: https://issues.apache.org/jira/browse/KARAF-7051

      Last tested and working version for me was 4.3.1. Not tested any other 4.3 versions.

      This is from test client:

      karaf@root()> date
      Jun Mon 17 14:41:50 2024
      karaf@root()> date
      Jun Mon 17 14:50:08 2024
      karaf@root(config)> config:list | grep ssh                                                                                                                                             
         ssh = org.apache.karaf.shell.ssh
         featuresBoot = instance/4.4.6, package/4.4.6, log/4.4.6, ssh/4.4.6, framework/4.4.6, system/4.4.6, eventadmin/4.4.6, feature/4.4.6, shell/4.4.6, management/4.4.6, service/4.4.6, jaas/4.4.6, deployer/4.4.6, diagnostic/4.4.6, wrap/2.6.14, bundle/4.4.6, config/4.4.6, kar/4.4.6
         sshHost = 0.0.0.0
         sshIdleTimeout = 90000
         sshPort = 8101
         sshRealm = karaf
         sshRole = ssh
         log4j2.logger.sshd.level = INFO
         log4j2.logger.sshd.name = org.apache.sshd
         feature.ssh = ssh
       

      client should have timed out before 2nd date command, but it didn't.

      We only noticed that because our containers started running out of memory and it was caused by many "client" instances open within container.

      Attachments

        Issue Links

        Activity

          This comment will be Viewable by All Users Viewable by All Users
          Cancel

          People

            jbonofre Jean-Baptiste Onofré
            nannou9 Piotr Klimczak
            Votes:
            0 Vote for this issue
            Watchers:
            2 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved:

              Slack

                Issue deployment