Some Eclipse Foundation services are deprecated, or will be soon. Please ensure you've read this important communication.
Bug 175264 - Agent Controller will not start on Linux.
Summary: Agent Controller will not start on Linux.
Status: CLOSED DUPLICATE of bug 131423
Alias: None
Product: z_Archived
Classification: Eclipse Foundation
Component: TPTP (show other bugs)
Version: unspecified   Edit
Hardware: PC Linux
: P1 blocker (vote)
Target Milestone: ---   Edit
Assignee: Igor Alelekov CLA
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks: 161096
  Show dependency tree
 
Reported: 2007-02-23 08:02 EST by Paul Slauenwhite CLA
Modified: 2016-05-05 10:51 EDT (History)
8 users (show)

See Also:


Attachments
servicelog.log (1.97 KB, application/x-gzip)
2007-03-29 20:04 EDT, Paul Slauenwhite CLA
no flags Details
servicelog.log (TPTP-4.4.0-200703270100C) (1.96 KB, application/x-gzip)
2007-04-04 11:34 EDT, Paul Slauenwhite CLA
no flags Details
servicelog.log (TPTP-4.4.0-200704261104) (2.03 KB, application/x-gzip)
2007-04-27 08:34 EDT, Paul Slauenwhite CLA
no flags Details
servicelog.log (TPTP-4.4.0-200705080100B) (1.96 KB, application/x-gzip)
2007-05-16 13:07 EDT, Paul Slauenwhite CLA
no flags Details
servicelog.log (TPTP-4.4.0-200705080100C and Sun 1.5.0.11 JRE) (2.13 KB, application/x-gzip)
2007-05-17 13:02 EDT, Paul Slauenwhite CLA
no flags Details
shmtest_32 (6.50 KB, application/octet-stream)
2007-05-18 01:59 EDT, Igor Alelekov CLA
no flags Details
updated library (18.94 KB, application/octet-stream)
2007-05-21 02:25 EDT, Igor Alelekov CLA
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description Paul Slauenwhite CLA 2007-02-23 08:02:00 EST
Agent Controller will not start on Linux.

Using the TPTP-4.4.0-200702211545 build, the Agent Controller for Linux IA32 will not start:

Starting Agent Controller.
Error starting transport layers, Agent controller exiting.
See servicelog.log for error report.
ACServer failed to start.

Servicelog.log:
<CommonBaseEvent creationTime="2007-02-23T12:46:57.355788Z" globalInstanceId="A5DEE24100056DC43060EB925DE3C903" msg="Unable to create shared memory: acbuffer." severity="50" version="1.0.1">
        <sourceComponentId component="AgentController" componentIdType="TPTPComponent" executionEnvironment="SharedMemListener.c, line 222" instanceId="1003" location="paules.site" locationType="IPV4" processId="5677" subComponent="Shared Memory TL" threadId="5677" componentType="Eclipse_TPTP"/>
        <situation categoryName="ReportSituation">
                <situationType xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:type="ReportSituation" reasoningScope="INTERNAL" reportCategory="LOG"/>
        </situation>
</CommonBaseEvent>
<CommonBaseEvent creationTime="2007-02-23T12:46:57.356244Z" globalInstanceId="A5DEE24100056F910E54F3FC1729C0A3" msg="An error was returned from TransportLayer(1003)::startTransportLayer errNum = -3" severity="50" version="1.0.1">
        <sourceComponentId component="AgentController" componentIdType="TPTPComponent" executionEnvironment="ConnectionManager.c, line 263" instanceId="2" location="paules.site" locationType="IPV4" processId="5677" subComponent="Connection Manager" threadId="5677" componentType="Eclipse_TPTP"/>
        <situation categoryName="ReportSituation">
                <situationType xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:type="ReportSituation" reasoningScope="INTERNAL" reportCategory="LOG"/>
        </situation>
</CommonBaseEvent>

OS:
Linux paules 2.6.16.21-0.8-smp #1 SMP Mon Jul 3 18:25:39 UTC 2006 x86_64 x86_64 x86_64 GNU/Linux

JRE:
java version "1.5.0"
Java(TM) 2 Runtime Environment, Standard Edition (build pxi32dev-20061002a (SR3) )
IBM J9 VM (build 2.3, J2RE 1.5.0 IBM J9 2.3 Linux x86-32 j9vmxi3223-20061001 (JIT enabled)
J9VM - 20060915_08260_lHdSMR
JIT  - 20060908_1811_r8
GC   - 20060906_AA)
JCL  - 20061002
Comment 1 Paul Slauenwhite CLA 2007-02-23 11:35:45 EST
This problem occurred on an EM64T machine running the following:

32 bit IBM JRE 1.5 SR3
32 bit Linux Agent Controller.

Interestingly, the Agent Controller starts using the following: 

64 bit (EM64T) IBM JRE 1.5 SR4
64 bit (EM64T) Linux Agent Controller.
Comment 2 Paul Slauenwhite CLA 2007-02-23 11:48:11 EST
(In reply to comment #1)

However, when launching a test using the 64-bit Agent Controller and JRE, I get the following exception:

Exception in thread "org.eclipse.hyades.execution.core.task.ReusableThread@c240c24" java.lang.NullPointerException
        at org.eclipse.hyades.execution.local.util.AgentControllerDescriptor.versionIsAtLeast(AgentControllerDescriptor.java:67)
        at org.eclipse.hyades.execution.local.util.AgentControllerDescriptor.isVersionAtLeast(AgentControllerDescriptor.java:194)
        at org.eclipse.hyades.execution.local.file.FileManagerFactory$3.run(FileManagerFactory.java:414)
        at org.eclipse.hyades.execution.core.task.ReusableThread.run(ReusableThread.java:131)
        at java.lang.Thread.run(Thread.java:803)
Exception in thread "org.eclipse.hyades.execution.core.task.ReusableThread@18951895" java.lang.NullPointerException
        at org.eclipse.hyades.execution.local.util.AgentControllerDescriptor.versionIsAtLeast(AgentControllerDescriptor.java:67)
        at org.eclipse.hyades.execution.local.util.AgentControllerDescriptor.isVersionAtLeast(AgentControllerDescriptor.java:194)
        at org.eclipse.hyades.execution.local.file.FileManagerFactory$3.run(FileManagerFactory.java:414)
        at org.eclipse.hyades.execution.core.task.ReusableThread.run(ReusableThread.java:131)
        at java.lang.Thread.run(Thread.java:803)
Exception in thread "org.eclipse.hyades.execution.core.task.ReusableThread@7de87de8" java.lang.NullPointerException
        at org.eclipse.hyades.execution.local.util.AgentControllerDescriptor.versionIsAtLeast(AgentControllerDescriptor.java:67)
        at org.eclipse.hyades.execution.local.util.AgentControllerDescriptor.isVersionAtLeast(AgentControllerDescriptor.java:194)
        at org.eclipse.hyades.execution.local.file.FileManagerFactory$3.run(FileManagerFactory.java:414)
        at org.eclipse.hyades.execution.core.task.ReusableThread.run(ReusableThread.java:131)
        at java.lang.Thread.run(Thread.java:803)
Exception in thread "org.eclipse.hyades.execution.core.task.ReusableThread@28e828e8" java.lang.NullPointerException
        at org.eclipse.hyades.execution.local.util.AgentControllerDescriptor.versionIsAtLeast(AgentControllerDescriptor.java:67)
        at org.eclipse.hyades.execution.local.util.AgentControllerDescriptor.isVersionAtLeast(AgentControllerDescriptor.java:194)
        at org.eclipse.hyades.execution.local.file.FileManagerFactory$3.run(FileManagerFactory.java:414)
        at org.eclipse.hyades.execution.core.task.ReusableThread.run(ReusableThread.java:131)
        at java.lang.Thread.run(Thread.java:803)
Exception in thread "org.eclipse.hyades.execution.core.task.ReusableThread@7ee87ee8" java.lang.NullPointerException
        at org.eclipse.hyades.execution.local.util.AgentControllerDescriptor.versionIsAtLeast(AgentControllerDescriptor.java:67)
        at org.eclipse.hyades.execution.local.util.AgentControllerDescriptor.isVersionAtLeast(AgentControllerDescriptor.java:194)
        at org.eclipse.hyades.execution.local.file.FileManagerFactory$3.run(FileManagerFactory.java:414)
        at org.eclipse.hyades.execution.core.task.ReusableThread.run(ReusableThread.java:131)
        at java.lang.Thread.run(Thread.java:803)
Exception in thread "org.eclipse.hyades.execution.core.task.ReusableThread@468a468a" java.lang.NullPointerException
        at org.eclipse.hyades.execution.local.util.AgentControllerDescriptor.versionIsAtLeast(AgentControllerDescriptor.java:67)
        at org.eclipse.hyades.execution.local.util.AgentControllerDescriptor.isVersionAtLeast(AgentControllerDescriptor.java:194)
        at org.eclipse.hyades.execution.local.file.FileManagerFactory$3.run(FileManagerFactory.java:414)
        at org.eclipse.hyades.execution.core.task.ReusableThread.run(ReusableThread.java:131)
        at java.lang.Thread.run(Thread.java:803)
Exception in thread "org.eclipse.hyades.execution.core.task.ReusableThread@70857085" java.lang.NullPointerException
        at org.eclipse.hyades.execution.local.util.AgentControllerDescriptor.versionIsAtLeast(AgentControllerDescriptor.java:67)
        at org.eclipse.hyades.execution.local.util.AgentControllerDescriptor.isVersionAtLeast(AgentControllerDescriptor.java:194)
        at org.eclipse.hyades.execution.local.file.FileManagerFactory$3.run(FileManagerFactory.java:414)
        at org.eclipse.hyades.execution.core.task.ReusableThread.run(ReusableThread.java:131)
        at java.lang.Thread.run(Thread.java:803)
Comment 3 Paul Slauenwhite CLA 2007-02-23 11:52:55 EST
(In reply to comment #2)

Opened defect https://bugs.eclipse.org/bugs/show_bug.cgi?id=175313.
Comment 4 Dave Smith CLA 2007-02-23 16:27:42 EST
I experienced the same failure and servicelog.log messages when trying to start the Linux IA32 Agent Controller (Runtime package) from TPTP-4.4.0-200702211545 build on a 32 bit machines running the following OS/JRE combinations:

1) SuSE SLES9 
Linux blackhawk-sles9 2.6.5-7.97-default #1 Fri Jul 2 14:21:59 UTC 2004 i686 i686 i386 GNU/Linux

java version "1.5.0"
Java(TM) 2 Runtime Environment, Standard Edition (build pxi32dev-20060511 (SR2))
IBM J9 VM (build 2.3, J2RE 1.5.0 IBM J9 2.3 Linux x86-32 j9vmxi3223-20060504 (JIT enabled)
J9VM - 20060501_06428_lHdSMR
JIT  - 20060428_1800_r8
GC   - 20060501_AA)
JCL  - 20060511a  

2)Red Hat Linux Advanced Server release 2.1AS/i686 (Pensacola)
Linux paradox-rh21as 2.4.9-e.49 #1 Fri Aug 6 11:56:52 EDT 2004 i686 unknown

java version "1.5.0_05"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_05-b05)
Java HotSpot(TM) Client VM (build 1.5.0_05-b05, mixed mode)
Comment 5 Igor Alelekov CLA 2007-03-05 09:48:42 EST
Paul,

I successfully launched AC on Suse SLES9 (x86_64) with AC for IA32 and IBM 32 bit jvm 1.5
The problem could be in leaving of unreleased shared memory buffers after AC termination (bug #173319).

Please repeat AC launch and if you get the same failure,
could you submit an output of "ipcs -m" command.
Comment 6 Igor Alelekov CLA 2007-03-13 07:19:06 EDT

*** This bug has been marked as a duplicate of bug 173319 ***
Comment 7 Paul Slauenwhite CLA 2007-03-29 20:03:49 EDT
This defect is blocking the 4.4 i2 TP1 (TPTP-4.4.0-200703270100A).

The Agent Controller (Linx-IA32) will not start with the following output:

ractest@paules:~/rac/bin> ./RAStart.sh
Starting Agent Controller.
Error starting transport layers, Agent controller exiting.
See servicelog.log for error report.
ACServer failed to start.

See the enclosed servicelog.log (DEBUG/CBE logging).

OS:

Linux paules 2.6.16.27-0.9-smp #1 SMP Tue Feb 13 09:35:18 UTC 2007 x86_64 x86_64 x86_64 GNU/Linux

JRE:

java version "1.5.0"
Java(TM) 2 Runtime Environment, Standard Edition (build pxi32dev-20070201 (SR4))
IBM J9 VM (build 2.3, J2RE 1.5.0 IBM J9 2.3 Linux x86-32 j9vmxi3223-20070201 (JIT enabled)
J9VM - 20070131_11312_lHdSMR
JIT  - 20070109_1805ifx1_r8
GC   - 200701_09)
JCL  - 20070126

Note, this is an EMT64T nmachine, but running a 32-bit JRE/Agent Controller.

This does not appear to be the result of unreleased shared memory buffers since the Agent Controller would not start after a new install of the OS.

ractest@paules:~/rac/bin> ipcs -m

------ Shared Memory Segments --------
key        shmid      owner      perms      bytes      nattch     status
0x00000000 393217     ractest   600        393216     2          dest
0x00000000 425986     ractest   600        393216     2          dest
0x00000000 229379     ractest   600        393216     2          dest
0x00000000 458756     ractest   600        393216     2          dest
0x00000000 491525     ractest   600        393216     2          dest
0x00000000 524294     ractest   600        393216     2          dest
0x00000000 557063     ractest   600        393216     2          dest
0x00000000 589832     ractest   600        393216     2          dest
0x00000000 688137     ractest   600        393216     2          dest
0x00000000 720906     ractest   600        393216     2          dest
0x00000000 1081355    ractest   600        393216     2          dest
0x00000000 1114124    ractest   600        12288      2          dest
0x00000000 1146893    ractest   600        393216     2          dest
0x00000000 1310734    ractest   600        393216     2          dest
Comment 8 Paul Slauenwhite CLA 2007-03-29 20:04:18 EDT
Created attachment 62443 [details]
servicelog.log
Comment 9 Paul Slauenwhite CLA 2007-04-02 10:41:04 EDT
Could this be caused by https://bugs.eclipse.org/bugs/show_bug.cgi?id=180050?  I will confirm afternoon.
Comment 10 Paul Slauenwhite CLA 2007-04-04 11:33:04 EDT
Using the TPTP-4.4.0-200703270100C build, this problem still exists.  Attaching service log.
Comment 11 Paul Slauenwhite CLA 2007-04-04 11:34:07 EDT
Created attachment 62944 [details]
servicelog.log (TPTP-4.4.0-200703270100C)
Comment 12 Paul Slauenwhite CLA 2007-04-04 14:17:01 EDT
Since this defect can only be reproduced on an EM64T machine and not experienced when testing using on 32-bit machines (e.g. RetHat), it is not blocking 4.4 i2.  As such, reducing the severity to critical.  

That said, this defect is still important for 4.4 since it was reproduced on the EM64T (SLES 10) using a 32-bit VM with the 32-bit OS extensions.  This is a common configuration that should be supported. 
Comment 13 jkubasta CLA 2007-04-04 14:18:15 EDT
targeting fix to i3
Comment 14 jkubasta CLA 2007-04-10 11:09:29 EDT
Igor has done some investigation and believes that this problem may be caused by the need to ship some 32-bit libraries with AC on Linux 64-bit.
These libraries are contained in the PI agent, perfmon and probekit plugins.
He believes these libraries should be rebuilt/ported to 64-bit. 

Valentina, Samson: can you confirm this is a requirement for piAgent and probekit?

Comment 15 Igor Alelekov CLA 2007-04-10 12:07:22 EDT
Hi Joanna,
It seems this is a missprint.
Porting of the PI agent, perfmon and probekit plugins is required for the bug #157486.
Comment 16 Geoffrey Hung CLA 2007-04-12 10:35:49 EDT
We have some problem using Agent COntroller on RHEL 5.0
Do we need to use 64bit Java on this platform?

OS: (running uname -a)
Linux lsc613 2.6.18-8.el5 #1 SMP Fri Jan 26 14:15:21 EST 2007 i686 i686 i386 GNU/Linux

JVM: (running java -version)
java version "1.5.0"
Java(TM) 2 Runtime Environment, Standard Edition (build pxi32devifx-20070323 (ifix 117677: SR4 + 116644 + 114941 + 116110 + 114881))
IBM J9 VM (build 2.3, J2RE 1.5.0 IBM J9 2.3 Linux x86-32 j9vmxi3223ifx-20070323 (JIT enabled)
J9VM - 20070322_12058_lHdSMR
JIT  - 20070109_1805ifx3_r8
GC   - WASIFIX_2007)
JCL  - 20070126
Comment 17 Kiryl Kazakevich CLA 2007-04-12 10:49:40 EDT
According to output of 'uname -a', your system is 32-bit (i686 means 6th generation of IA-32). You need 32-bit agent controller and 32-bit Java.
Comment 18 Paul Slauenwhite CLA 2007-04-27 08:31:11 EDT
Using the TPTP-4.4.0-200704261104 build, the 32-bit Agent Controller running on the 32-bit IBM Java 1.5 SR4 JVM could not be started.  See enclosed service log.
Comment 19 Paul Slauenwhite CLA 2007-04-27 08:34:02 EDT
Created attachment 65197 [details]
servicelog.log (TPTP-4.4.0-200704261104)
Comment 20 jkubasta CLA 2007-04-27 08:41:05 EDT
Alan, what JVM and build did you succeed with?
Comment 21 Alan Haggarty CLA 2007-04-27 09:36:25 EDT
I used:
agntctrl.linux_ia32-TPTP-4.4.0-200703270100D.zip

and
java version "1.5.0"
Java(TM) 2 Runtime Environment, Standard Edition (build pxi32dev-20060511
IBM J9 VM (build 2.3, J2RE 1.5.0 IBM J9 2.3 Linux x86-32 j9vmxi3223-200605T enabled)
J9VM - 20060501_06428_lHdSMR
JIT  - 20060428_1800_r8
GC   - 20060501_AA)
JCL  - 20060511a

all are 32 bit and had no problems. Probably have nothing to do with Paul's
original problem but maybe clears up some of the other appends.
Comment 22 Paul Slauenwhite CLA 2007-04-30 07:45:16 EDT
(In reply to comment #21)

Alan, what SR is this?  I would have thought that using a new SR would resolve this type of issue, not create it:).  I will give your configuration a try to see if it helps diagnose the problem.
Comment 23 Alan Haggarty CLA 2007-05-03 09:15:57 EDT
Sorry, that was cut off above. My java was SR2.

It is in /usr/lib/ibm-java2-i386-50 on blackhawk-sles9
Comment 24 Paul Slauenwhite CLA 2007-05-16 13:06:22 EDT
Using the TPTP-4.4.0-200705080100B I am still seeing this problem with both the IBM 1.5 SR2 and SR4a JREs (see attached servicelog.log in DEBUG mode when attempting to start the Agent Controller using the IBM 1.5 SR2 JRE.

Note: This defect is blocking Test Project testing.
Comment 25 Paul Slauenwhite CLA 2007-05-16 13:07:13 EDT
Created attachment 67451 [details]
servicelog.log (TPTP-4.4.0-200705080100B)
Comment 26 Paul Slauenwhite CLA 2007-05-16 13:08:31 EDT
Can we increase the severity of this defect to blocker since there is no work-around and it is blocking most TPTP fucntion when using this configuration.
Comment 27 jkubasta CLA 2007-05-16 13:15:32 EDT
Igor, please make this top priority
Comment 28 Igor Alelekov CLA 2007-05-17 08:40:52 EDT
Paul,
could you tell what limits for shared memory does your system have?
It could be got with "ipcs -lm" command.
Comment 29 Samson Wai CLA 2007-05-17 10:44:30 EDT
When I first start the 32-bit AC on Linux/EM64T running Fedora Core 6, I get this error:
ACServer: error while loading shared libraries: libstdc++-libc6.2-2.so.3: cannot open shared object file: No such file or directory

Then I have to install the following missing libraries from the FC6 installation CD:
compat-libstdc++-296-2.96-138.i386.rpm

After that the 32-bit AC starts up fine and I can run a profiling scenario using JVMTI successfully.

My Linux version is:
Linux tptp64lnx1.torolab.ibm.com 2.6.18-1.2798.fc6 #1 SMP Mon Oct 16 14:39:22 EDT 2006 x86_64 x86_64 x86_64 GNU/Linux

And for some reason non of the IBM JRE can be run on my machine so I end up using a 32-bit Sun JRE:
java version "1.5.0_11"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_11-b03)
Java HotSpot(TM) Client VM (build 1.5.0_11-b03, mixed mode, sharing)

My "ipcs -lm" output looks like:
------ Shared Memory Limits --------
max number of segments = 4096
max seg size (kbytes) = 32768
max total shared memory (kbytes) = 8388608
min seg size (bytes) = 1
Comment 30 Paul Slauenwhite CLA 2007-05-17 12:19:34 EDT
(In reply to comment #28)
> Paul,
> could you tell what limits for shared memory does your system have?
> It could be got with "ipcs -lm" command.
> 

ractest@paules:~> ipcs -lm

------ Shared Memory Limits --------
max number of segments = 4096
max seg size (kbytes) = 18014398509481983
max total shared memory (pages) = 1152921504606846720
min seg size (bytes) = 1
Comment 31 Paul Slauenwhite CLA 2007-05-17 13:01:14 EDT
(In reply to comment #29)

Using the same Sun 1.5 (11) JRE, the problem persists:

ractest@paules:~> java -version
java version "1.5.0_11"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_11-b03)
Java HotSpot(TM) Client VM (build 1.5.0_11-b03, mixed mode)
ractest@paules:~> cd rac/bin
ractest@paules:~/rac/bin> ./SetConfig.sh
Specify the fully qualified path of "java" (e.g. /usr/java1.4/jre/bin/java):
  Default>"/home/ractest/SUN_JRE_1_5_0_11_IA32/jre/bin/java" (Press <ENTER> to accept the default value)
  New value>
Network access mode (ALL=allow any host, LOCAL=allow only this host, CUSTOM=list of hosts):
  Default>"LOCAL" (Press <ENTER> to accept the default value)
  New value>ALL
Security enabled. (true/false):
  Default>"FALSE" (Press <ENTER> to accept the default value)
  New value>
Please enter the JBoss Application Server Home:
  Default>"" (Press <ENTER> to accept the default value)
  New value>
Please enter the JOnAS Application Server Home:
  Default>"" (Press <ENTER> to accept the default value)
  New value>

ractest@paules:~/rac/bin> ./RAStart.sh
Starting Agent Controller.
Error starting transport layers, Agent controller exiting.
See servicelog.log for error report.
ACServer failed to start.

Is there any way you can istall SLES 10 on your x86 Linux machine? 
Comment 32 Paul Slauenwhite CLA 2007-05-17 13:02:02 EDT
Created attachment 67679 [details]
servicelog.log (TPTP-4.4.0-200705080100C and Sun 1.5.0.11 JRE)
Comment 33 Igor Alelekov CLA 2007-05-17 13:28:11 EDT
(In reply to comment #31)
> Is there any way you can istall SLES 10 on your x86 Linux machine? 
Today I successfully launched 32-bit AC on SLES10 on EM64T machine. 
Comment 34 Igor Alelekov CLA 2007-05-17 14:11:25 EDT
(In reply to comment #30)
> ractest@paules:~> ipcs -lm
> ------ Shared Memory Limits --------
> max number of segments = 4096
> max seg size (kbytes) = 18014398509481983
> max total shared memory (pages) = 1152921504606846720
> min seg size (bytes) = 1

Paul, these digits are so strange. How much memory has you system?
What does command "sysctl kernel.shmmax" say?
Comment 35 Paul Slauenwhite CLA 2007-05-17 14:24:06 EDT
(In reply to comment #34)

I do not have "sysctl" command.

Comment 36 Paul Slauenwhite CLA 2007-05-17 14:46:46 EDT
(In reply to comment #33)
> (In reply to comment #31)
> > Is there any way you can istall SLES 10 on your x86 Linux machine? 
> Today I successfully launched 32-bit AC on SLES10 on EM64T machine. 
> 
 
Great news!  What is the difference?  What TPTP version are you running?  Are you running service pack 1?  My machine has a bare bones install of SLES 10.
Comment 37 Igor Alelekov CLA 2007-05-17 14:56:54 EDT
> Great news!  What is the difference?  What TPTP version are you running?  Are
> you running service pack 1?  My machine has a bare bones install of SLES 10.

I used AC build from May 14th and also my own build frum the current CVS code.
Now I don't remember all details about my system and will inform you tomorrow.

Paul, what about memory on your system?
Probably your system (8-10 processors?) could have special characteristics cousing this defect.


Comment 38 Paul Slauenwhite CLA 2007-05-17 15:20:13 EDT
(In reply to comment #37)

My machine is an IBM 8212 K4U with 2048 MB of memory and 1 x Pentium D 930 @ 3000MHz.
Comment 39 Igor Alelekov CLA 2007-05-18 00:58:16 EDT
(In reply to comment #35)
> (In reply to comment #34)
> I do not have "sysctl" command.

Please repeat the command as root
Comment 40 Igor Alelekov CLA 2007-05-18 01:59:34 EDT
Created attachment 67766 [details]
shmtest_32

Paul, it seems that this defect depends on the #131423.
Could you execute the attached shmtest_32 program on your system and submit an output?
Comment 41 Paul Slauenwhite CLA 2007-05-18 07:04:46 EDT
(In reply to comment #39)
> (In reply to comment #35)
> > (In reply to comment #34)
> > I do not have "sysctl" command.
> 
> Please repeat the command as root
> 

Sorry.  I did not realize you needed to be root to execute this command:

paules:/home/ractest # sysctl kernel.shmmax
kernel.shmmax = 18446744073709551615
Comment 42 Paul Slauenwhite CLA 2007-05-18 07:06:38 EDT
(In reply to comment #40)
> Created an attachment (id=67766) [details]
> shmtest_32
> 
> Paul, it seems that this defect depends on the #131423.
> Could you execute the attached shmtest_32 program on your system and submit an
> output?
> 
 
[As root]

paules:/home/ractest/Desktop # ./shmtest_32

----- Shared Memory Limits: -----
Max number of segments: 4096
Max segment size (kbytes): 4194303
Total shared memory (pages): 4294967040
Min segment size (bytes): 1

----- Kernel Info: -----
SHMMAX: -1

----- Creating shared memory segment of 1024 kbytes -----
Success

----- Creating shared memory segment of 512 kbytes -----
Success

----- Creating shared memory segment of 128 kbytes -----
Success
Comment 43 Igor Alelekov CLA 2007-05-18 07:09:52 EDT
> paules:/home/ractest/Desktop # ./shmtest_32
> ----- Shared Memory Limits: -----
> Max number of segments: 4096
> Max segment size (kbytes): 4194303
> Total shared memory (pages): 4294967040
> Min segment size (bytes): 1
> ----- Kernel Info: -----
> SHMMAX: -1
> ----- Creating shared memory segment of 1024 kbytes -----
> Success
> ----- Creating shared memory segment of 512 kbytes -----
> Success
> ----- Creating shared memory segment of 128 kbytes -----
> Success

Great!
It means that this defect will be resolved with the patch 131423 check in.
Comment 44 Paul Slauenwhite CLA 2007-05-18 07:13:20 EDT
(In reply to comment #43)
> > paules:/home/ractest/Desktop # ./shmtest_32
> > ----- Shared Memory Limits: -----
> > Max number of segments: 4096
> > Max segment size (kbytes): 4194303
> > Total shared memory (pages): 4294967040
> > Min segment size (bytes): 1
> > ----- Kernel Info: -----
> > SHMMAX: -1
> > ----- Creating shared memory segment of 1024 kbytes -----
> > Success
> > ----- Creating shared memory segment of 512 kbytes -----
> > Success
> > ----- Creating shared memory segment of 128 kbytes -----
> > Success
> 
> Great!
> It means that this defect will be resolved with the patch 131423 check in.
> 

Fantastic!  I have made this defect dependent on 131423.  

Joanna, can we increase the severity of 131423 since it is blocking testing on Linux?
Comment 45 Igor Alelekov CLA 2007-05-21 02:25:17 EDT
Created attachment 67943 [details]
updated library

Paul, this is the fixed library for Linux IA32 AC.
To test it please copy the library to the $AC_HOME/lib directory.
I used last AC build agntctrl.linux_ia32-TPTP-4.4.0-200705200100.
Comment 46 Paul Slauenwhite CLA 2007-05-23 12:17:39 EDT
(In reply to comment #45)
> Created an attachment (id=67943) [details]
> updated library
> 
> Paul, this is the fixed library for Linux IA32 AC.
> To test it please copy the library to the $AC_HOME/lib directory.
> I used last AC build agntctrl.linux_ia32-TPTP-4.4.0-200705200100.
> 

We have a winner!  Using this library, I can start the Agent Controller and launch a test.  Returning this defect as a dup of 131423.  Thanks Igor for all your help.

I am assumming the debug statements are going to be removed:

ractest@paules:~/rac/bin> ./RAStart.sh
Starting Agent Controller.
shm -1
validated 1048576
ACServer started successfully.
ractest@paules:~/rac/bin>

*** This bug has been marked as a duplicate of bug 131423 ***
Comment 47 Paul Slauenwhite CLA 2007-06-02 12:33:14 EDT
Verified in TPTP-4.4.0-200705310943.