Bug 3080

Summary: errors in JUnit tests in frysk-core (libelf problem?)
Product: frysk Reporter: Stepan Kasal <skasal>
Component: generalAssignee: Tim Moore <timoore>
Severity: normal CC: qiyaoltc, woodzltc, zhengyzy
Priority: P2    
Version: unspecified   
Target Milestone: ---   
Host: Target:
Build: Last reconfirmed:
Bug Depends on:    
Bug Blocks: 1553    
Attachments: the change

Description Stepan Kasal 2006-08-17 10:04:52 UTC
There is a dozen of errors when running JUnit tests under frysk-core.  They all
look similar; here is a typical example:

got TaskException
   at frysk.proc.LinuxHost.sendCreateAttachedProc(TestRunner)
   at frysk.proc.HostState$1.handleCreateAttachedProc(TestRunner)
   at frysk.proc.Host$2.execute(TestRunner)
   at frysk.event.EventLoop.runEventLoop(TestRunner)
   at frysk.event.EventLoop.runPolling(TestRunner)
   at frysk.proc.TestLib.assertRunUntilStop(TestRunner)
   at frysk.proc.TestLib.assertRunUntilStop(TestRunner)
   at frysk.proc.TestRun.testCreateAttachedContinuedProc(TestRunner)
   at frysk.junit.Runner.<init>(TestRunner)
   at TestRunner.main(TestRunner)
Caused by: frysk.proc.TaskFileException: Could not open /proc/17045/exe for reading
   at frysk.proc.IsaFactory.getIsa(TestRunner)
   at frysk.proc.IsaFactory.getIsa(TestRunner)
   at frysk.proc.LinuxTask.sendrecIsa(TestRunner)
   at frysk.proc.Task.getIsa(TestRunner)
   at frysk.proc.LinuxTask.setupMapsXXX(TestRunner)
   at frysk.proc.LinuxTask.<init>(TestRunner)
   at frysk.proc.LinuxHost.sendCreateAttachedProc(TestRunner)
   ...22 more
Caused by: lib.elf.ElfFileException: Could not open /proc/17045/exe for reading
   at lib.elf.Elf.elf_begin(TestRunner)
   at lib.elf.Elf.<init>(TestRunner)
   at frysk.proc.IsaFactory.getIsa(TestRunner)
   ...28 more
Comment 1 Mark Wielaard 2006-08-18 15:27:52 UTC
This is caused by:

        2006-08-16  Tim Moore  <timoore@redhat.com>
        * LinuxHost.java (sendCreateAttachedProc,
        PollWaitOnSigChld.anonymous.forkEvent): Rethrow TaskException as
        RuntimeException instead of eating it.

Which I guess is a reasonable thing to do, but does break this testcase.

Tim, could you take a look?
Comment 2 Stepan Kasal 2006-08-21 08:54:00 UTC
Created attachment 1235 [details]
the change

This patch is the reversed version of the change which caused the bug.
I verified that applying it ``fixes'' (or covers?) the bug.

But I have no reason to belive that this is the way to go.
This is not a proposed patch, it is just an exhibit illustrating the Mark's
Comment 3 Wu Zhou 2006-08-21 10:39:12 UTC
The symptom is the same as what I reported in 3081.

What is the final solution?  I noticed there is no exception handling afer
applying the above patch.
Comment 5 Stepan Kasal 2006-08-21 10:56:28 UTC
*** Bug 3081 has been marked as a duplicate of this bug. ***
Comment 6 Stepan Kasal 2006-08-21 12:18:28 UTC
I have verified that the bug is indeed fixed.  Thanks.