Fix released (Re: CAUTION: Problem with areaDetector ImageJ PVA viewer)

[Ralph Lange <ralph.lange@gmx.de>]

Dear all,

On behalf of the EPICS Java developers I have published and released to Maven Central a bugfix version 4.3.2 of the epics-pvaclient code, which fixes the issue Mark described.

Maven projects that depend on the current snapshot version of the epics-core bundle (7.0.2-SNAPSHOT) will pull the update automatically. All other projects will have to add an explicit dependency to epics-pvaclient 4.3.2 or download that jar. See [1] for examples in different build systems.

Thanks a lot to Mark Rivers for reporting the issue and testing/verifying the fix and to Marty Kraimer for fixing the bug.

And, of course, sorry for the inconveniences caused.

Cheers,
~Ralph

[1] https://search.maven.org/artifact/org.epics/epics-pvaclient/4.3.2/jar

On Sun, 30 Sep 2018 at 00:28, Mark Rivers <rivers@cars.uchicago.edu> wrote:

Folks,

 

I have found a serious issue with the EPICS V4 Java client code which is exposed with the areaDetector ImageJ PVA viewer (EPICS_NTNDA_Viewer.java).

 

When the areaDetector IOC is not running the ImageJ plugin keeps trying to connect.  I have found that the broadcast search requests have an ever-increasing number of duplicate PVs in them.  The search requests normally only occur at 0.5 Hz or so.  However, as the number of duplicate PVs per search increases, each search requires multiple Ethernet packets.  After a few hours the number of broadcast packets is very large.  I have seen those broadcasts consume >90% of the CPU time on all of my VME IOCs.

 

It appears that the problem is not in the ImageJ plugin, but rather in the underlying epics-pvaclient Java code.  Marty Kraimer and I are planning to test a fix to that soon.

 

In the meantime I would discourage people from using the ImageJ EPICS_NTNDA_Viewer, unless you are sure to kill ImageJ when the areaDetector IOC is not running.  Other applications using the epics-pvaclient Java library may also have this problem.

 

Mark