That is the question: my Java application is running, so how to get the path/name to the java binary that is running it? For example, 'C:\Program Files\Java\jre6\bin\java.exe' in Windows or '/usr/bin/java' in Linux. I need a platform independent solution, preferably relying only on the JDK.
The context for the question is the following. I'm developing two Java applications, let's say FirstApplication and SecondApplication. The FirstApplication will start the SecondApplication using the ProcessBuilder API. The problem is that I need to tell the path to the Java binary to start the SecondApplication. For example,
String cmd = "C:\Program Files\Java\jre6\bin\java.exe -cp <second application classpath> com.secondapplication.MainClass";
Process p = new ProcessBuilder(cmd).start();
Possible alternative solutions to this could be assume that there is a java binary available through the PATH system variable, or assume that the JAVA_HOME system variable is set. But this may not be always the case. Also, the final user of the application (targeting to other Java developers, as well as final users) may have more than one JVM installed on his system, and may execute the FirstApplication expliciting what JVM he wants. Ideally, the SecondAplication should run using the exact same Java binary as the FirstApplication, without relying on such system variables.
Thank you
Related
Testing Arquillian 1.9.final TOMCAT-EMBED-7 container, and I'm getting questionable results around creating a WebArchive for testing.
In /src/main/resources, I have several configuration files that I do not want to use when running the integration tests, instead I want to provide named ones stored in /src/embed-itest/resources.
org.jboss.shrinkwrap.api.Filter x = Filters.exclude(".*Test.*|.*xml|.*properties");
WebArchive webArchive = ShrinkWrap
.create(WebArchive.class, "mytest.war")
.addPackages(true, x, "com.myapp")
//and some other additions
Then at the end of the srhinkwrap process, I add the specific test files I want to use:
File n = new File("src/embed-itest/resources/test-log4j.properties");
webArchive.addAsResource(n,"log4j.properties");
However, the behavior is still running as though it is using the /src/main/resources/log4j.properties. I've verified the _DEFAULT_DEFAULT_mytest.war really does have the test-log4j.properties content as log4j.properties, but running the tests the behavior is that of /src/main/resources/log4j.properties. (and this is true for other configuration files, such as camelContext.xml I've tried to override).
Anyone have some insight please? I was hoping to leverage the ability to create a custom WebArchive with specific files in the archive to more precisely test, but the actual behavior seems to be as if it was the 'standard' created war limiting what I thought was a great capability of arquillian.
I think the problem is that yo are using the Tomcat embedded approach, which means you are sharing the JVM of your tests with your Tomcat instance. I suggest you try with managed or remote mode.
I am having a problem with using Apache/Perl to get access to Excel files using Microsoft Data Access Component (MDAC). Somehow I must set the "CommonProgramFiles(x86)" system environment variable in order to get this to work. Otherwise, I keep getting this error message:
System.InvalidOperationException: The .Net Framework Data Providers
require Microsoft Data Access Components(MDAC). Please install
Microsoft Data Access Components(MDAC) version 2.6 or later. --->
System.IO.FileNotFoundException: Retrieving the COM class factory for
component with CLSID {2206CDB2-19C1-11D1-89E0-00C04FD7A829} failed due
to the following error: 8007007e.
The server configuration is:
Windows Server 2008 R2 in 64-bit
Server is installed with Microsoft Access Database Engine 2010
Apache 2.2.25 (that is 32-bit)
Perl 5.12.3 v5 (that is also in 32-bit)
I have my Perl CGI script to call my C# program (that is built for "Any CPU").
The C# program uses MDAC to open and read Excel files (not trying to launch Excel, only try to read data from the Excel files).
I have verified that the server has the latest MDAC available in these 2 folders:
C:\Program Files\Common Files\System\Ole DB
C:\Program Files (x86)\Common Files\System\Ole DB
I have also checked the registries and they look fine. Anyway, I don't have any problem running my C# program directly at the command prompt (it can use MDAC to get access to Excel files). I only have the problem when I use Apache/Perl to use my Perl CGI script to call my C# program (that is when I get that error with MDAC).
I can work around this problem by specifying CommonProgramFiles(x86) in my Perl CGI script, like this:
$ENV{ "CommonProgramFiles(x86)" } = "C:\\Program Files (x86)\\Common Files";
I have this question:
Why do I have this problem? And why setting that CommonProgramFiles(x86) system environment variable can workaround this problem? Why that system environment variable is empty before I set it? Does this have to do with the fact that I am running 32-bit Apache/Perl in a Windows operating system that is 64-bit?
Please help me to understand this issue. Thanks in advance.
(The original version of this post had a question about a second problem. Turned out that problem had to do with an extra double quote in the string. I fixed this, and that problem has gone away. That's why I have removed that second question from the post)
Jay Chan
I did some more research and the issue is the following: until release 2.4.9, the Apache startup routines have mapped "commonprogramfiles(x86)" to "commonprogramfiles_x86_" and that variable does not exists in the environment unless you create it... I have not tested it, but creating that environment variable and making it point to the same location as commonprogramfiles(x86) would probably fix the issue too.
Since the compiled Apache distributions are only using up to version 2.4.46 as we speak, they don't have the fix that allows the parenthesis in environment variables. That's why you still need the PassEnv directive to ensure that Apache passes the correct values to 32-bit CGI scripts.
The following post has some useful details about this:
https://bz.apache.org/bugzilla/show_bug.cgi?id=46751
I used to have the same problem in with Apache 2.4 with dBase compiled apps using ADO-32 bit as dBase is 32-bit. Something has recently changed, possibly with Windows 10 2004 20H2. I needed this fix in July-Aug 2020 but now I don't, the environment variable already exists. Since my Apache version is dated April 2020, that cannot be the reason for the change.
I tried to do some research about this but all I could find is that those environment variables are system ones existing at least since 2017, so why I needed to set this var is a mystery to me, but I would like to understand this, so if you find something, post a follow-up...
https://learn.microsoft.com/en-us/windows/deployment/usmt/usmt-recognized-environment-variables
I struggled on this issue for a few days but didn't get a right answer yet.
Here is the Problem Description:
I wrote a normal Java program (Program-A), and wrote a Windows-based native agent (*.dll, written in C/C++) with Agent_OnLoad, Agent_OnAttach, Agent_OnUnload method, which works fine if using Java command-line flag (-agentlib). Then I wrote another Java program to attach the native agent onto a runing the Java Program-A (see the below code piece for VM attach and loadAgentPath), however I got the exception:
com.sun.tools.attach.AgentLoadException: Failed to load agent library
I tried to change the agentPath (absolute or relative file path) this or that way, none of these works. Should I try some other way to make this work. What I need is to attach a native agent onto a runing java program rather than using command-line flag.
Does anyone know the root cause or a clue for the solution?
BTW, the command line to run attach VM Java code as:
java -Djava.library.path=D:\DevTools\Java7\jre7\bin -classpath .;./tools.jar com.xxx.TestAgentVMAttacher
...
VirtualMachine virtualMachine = com.sun.tools.attach.VirtualMachine.attach(pid); // Note: this code line is executed normally, I am sure the pid is correct
...
agentPath = theFilePath + "/myagent.dll"; // Note: I am sure the dll file path is correct
virtualMachine.loadAgentPath(agentPath,null); // Note: this code line would cause the exception (AgentLoadException) as I mentioned above, no matter how I set the agentPath, even I set it as null, same exception happened.
Environment related info:
- OS: Windows XP
- Java Version: Java(TM) SE Runtime Environment (build 1.7.0-b147)
Eventually I found the answer for my question, I had a wrong method name ('Agent_Attach') in Agent.cpp file, the correct one should be 'Agent_OnAttach', with this fix, my agent lib (.dll) can be loaded to a running Java program now.
Full story:
I am trying to start up an instance of hudson with a larger memory allocation and I'm currently using scripts owned by root that I can't modify directly to pass arguments. However the script currently passes the $JAVA_ARGS variable when starting up the service. I have exported the required parameters to JAVA_ARGS but the application still appears to be bound by the old memory restrictions.
Question:
Is there a way to find out which command line parameters were used to start up the instance. More specifically I'm looking to find the values that were passed (if any) to Xmx and Xms.
java version "1.6.0_12"
Java(TM) SE Runtime Environment (build 1.6.0_12-b04)
Java HotSpot(TM) 64-Bit Server VM (build 11.2-b01, mixed mode)
After some searching I came about a pretty simple solution (which I'm a little embarrassed to have missed for so long). You can see the command line to any command running in linux with ps, as long as you pass the correct flags. I just made a call to ps -fHu hudson and was able to see the full command line call to java which showed the passed in parameters.
Since you can export $JAVA_ARGS, maybe you can override $PATH to trick the script to run another program instead of the JVM, which could be a program that simply writes its arguments somewhere.
I have a third party library that I'm attempting to incorporate into a simulation. We have the static library (.a), along with all of it's runtime dependencies (shared objects). I've created a very simple application (in C) that is linked against the library. All it does is call an initialization function that is part of the third party library's API, and exits. When I run this directly from the command line, it works fine. If I submit the executable to our Condor grid, it fails with a seg fault on strncpy (libc.so.6). I've forced condor to only run the executable on a particular machine, and if I run it directly on that machine, it works fine.
I'm mostly a Java programmer... limited amount of native coding experience. I'm familiar with tools such as nm, ldd, catchsegv, etc... to the point where I can run them. I don't really know where to start looking for an issue though.
I've run ldd directly on the executing machine, and via a script submitted through condor, along with my executable. ldd reports the same files in both cases.
I don't understand how running it directly would work, but it would fail being run by condor. The process that ultimately executes the program, condor_startd, is a process that starts as root, and changes its effective uid to the submitter. Perhaps this has something to do with it?
Don't know why this would cause an issue, but the culprit was the LANG environment variable. It was not set when running under Condor, but was set to US_EN.UTF-8 when running locally. Adding this value to the condor execution environment fixed the problem.