JPPF Issue Tracker
star_faded.png
Please log in to bookmark issues
bug_report_small.png
CLOSED  Bug report JPPF-319  -  Master node is reporting an incorrect number of slaves
Posted Sep 02, 2014 - updated Aug 15, 2018
icon_info.png This issue has been closed with status "Closed" and resolution "RESOLVED".
Issue details
  • Type of issue
    Bug report
  • Status
     
    		 Closed
  • Assigned to
    lolo4j
  • Progress
       
  • Type of bug
    Not triaged
  • Likelihood
    Not triaged
  • Effect
    Not triaged
  • Posted by
    lolo4j
  • Owned by
    Not owned by anyone
  • Category
    Node
  • Resolution
    RESOLVED
  • Priority
    Normal
  • Reproducability
    Always
  • Severity
    Normal
  • Targetted for
    icon_milestones.png JPPF 4.2.x
Issue description
In some situations, a master node may report an inaccurate number of slaves nodes. I've seen this happen when provisioning N nodes, shutting down one of the slaves via JMX, then provisioning N slaves again. In this scenario, we end up with the master reporting N-1 nodes, whereas N are actually running.
Steps to reproduce this issue
  • start a driver and a node
  • start the admin console and provision 3 slave nodes for the running master ==> all 3 nodes start and the master reports 3 slaves
  • in the console, select one of the slaves and shut it down ==> the master now correctly reports 2 slaves
  • still in the console, request again the provisioning of 3 slaves ==> an additional slave is started, but the master still reports 2 slaves instead of 3

#3
Comment posted by
lolo4j
Sep 02, 20:23
The problem is in how the names of the slaves working directories are managed. Here the issue occurs because we start a new slaves with a working directory that is already in use, thus the map of directory names to slave process contains only 2 entries instead of 3 because one of the entries is overwritten.

#4
Comment posted by
lolo4j
Sep 02, 21:20
Fixed in: