How to identify exactly where exception occurred in Jenkins pipeline? - jenkins

In Jenkins pipeline build, sometimes I've seen null pointer or other exceptions like -
java.lang.NullPointerException: Cannot invoke method trim() on null object
Generally if we run Java program through IDE or command line, if an exception occurs we see at which line number the exception has occurred.
But with Jenkins build output console, it does not show the line number where the exception has occurred.
In this case, based on method name ie trim() from log, I check wherever trim() method is used. But as I've used it at multiple places in same method, it becomes difficult to identify exactly where error has occurred.
Another way is to add echo statements and re-run build and see where it gives this exception but this is time consuming.
Is there any better way/plugin using which I can identify at which line of pipeline code exception has occurred?


I don't really know if it's possible to show the exact line number, but you can wrap your code in try-catch statements and then show the exception info in the catch, like so:
try {
// line with trim()
catch (ex) {
println "Exception while trimming: $ex"
}

Related

How to get more information for an error generated in libcontainer/rootfs_linux.go?

Code used from github: https://github.com/opencontainers/runc/blob/master/libcontainer/rootfs_linux.go#L59
This is the if block to debug for our interest....
if err := mountToRootfs(m, config.Rootfs, config.MountLabel, hasCgroupns); err != nil {
return newSystemErrorWithCausef(err, "mounting %q to rootfs at %q", m.Source, m.Destination)
}
Here in the if condition, there is an error generated thru newSystemErrorWithCausef function and which internally uses generic_error.go, fmt.Sprintf to format the error string.
In our integration, only once it hit this if block. And in the process of getting more details, I wanted to troubleshoot and understand why this happened.
As of now tried using fmt.Printf(........) in line 55 and also just before if block of line 59.
And also tried putting above statement in very first line of prepareRootfs function. And then tried newSystemErrorWithCause(err, "some other message") in the first line of prepareRootfs function. Overall I dont see any information generated with my messages.
Please provide any information to understand more of this issue. Please note I am not looking for why this error might have been caused. I am looking for how to get more information from that function of prepareRootfs.

What does this ejabberd / erlang stack trace mean?

could someone help me figure out what to look for in this stacktrace? I've never touched Erlang, so I don't really know where to start.
We're using v19.09 of ejabberd, and getting this error log for many users:
2020-03-14 15:29:31.096 [error] <0.2029.48>#mod_http_api:handle:257 REST API Error: set_presence([{<<"user">>,<<"53265363">>},{<<"host">>,<<"chat.us.com">>},{<<"resource">>,<<"ssid-3465">>},{<<"type">>,<<"available">>},{<<"show">>,<<"chat">>},{<<"status">>,<<"{\"current_lobby_id\":\"none\",\"status\":\"{\\r\
\\t\\\"rich_presence\\\": \\\"{\\\\r\\\
\\\\t\\\\\\\"rich_presence\\\\\\\": \\\\\\\"?s=PlayerStatusInMap?p0=21?p1=2?p2=3\\\\\\\",\\\\r\\\
\\\\t\\\\\\\"session_id\\\\\\\": \\\\\\\"ssid-3465-38a44f60-9df7-4c79-8c8e-1cc99e5039dd\\\\\\\"\\\\r\\\
}\\\",\\r\
\\t\\\"current_lobby_id\\\": \\\"{\\\\r\\\
\\\\t\\\\\\\"lobby_id\\\\\\\": \\\\\\\"c791cfaa-9fd3-4d0d-9d5c-1aeb9efb12e2#hangouts.chat.us.com\\\\\\\",\\\\r\\\
\\\\t\\\\\\\"privacy\\\\\\\": \\\\\\\"friends_only\\\\\\\"\\\\r\\\
}\\\"\\r\
}\",\"region\":\"\"}">>},{<<"priority">>,<<"0">>}]) -> exit:{noproc,{p1_server,call,[none,{set_presence,{presence,<<>>,available,<<>>,{jid,<<"953265363">>,<<"chat.us.com">>,<<"oak/ssid-3465-1cc99e5039dd">>,<<"1953265363">>,<<"chat.us.com">>,<<"4c79-8c8e-1cc99e5039dd">>},{jid,<<"1953265363">>,<<"chat.us.com">>,<<>>,<<"1953265363">>,<<"chat.us.com">>,<<>>},chat,[{text,<<"en">>,<<"{\"current_lobby_id\":\"none\",\"status\":\"{\\r\
\\t\\\"rich_presence\\\": \\\"{\\\\r\\\
\\\\t\\\\\\\"rich_presence\\\\\\\": \\\\\\\"?s=PlayerStatusInMap?p0=21?p1=2?p2=3\\\\\\\",\\\\r\\\
\\\\t\\\\\\\"session_id\\\\\\\": \\\\\\\"ssid-348c8e-1cc99e5039dd\\\\\\\"\\\\r\\\
}\\\",\\r\
\\t\\\"current_lobby_id\\\": \\\"{\\\\r\\\
\\\\t\\\\\\\"lobby_id\\\\\\\": \\\\\\\"c791cfaa-9fd3-4d0d-9d5c-1aeb9efb12e2#hangouts.chat.us.com\\\\\\\",\\\\r\\\
\\\\t\\\\\\\"privacy\\\\\\\": \\\\\\\"friends_only\\\\\\\"\\\\r\\\
}\\\"\\r\
}\",\"region\":\"\"}">>}],0,[],#{}}},1000]}} [{p1_server,call,3,[{file,"src/p1_server.erl"},{line,210}]},{mod_http_api,handle2,4,[{file,"src/mod_http_api.erl"},{line,267}]},{mod_http_api,handle,4,[{file,"src/mod_http_api.erl"},{line,228}]},{mod_http_api,perform_call,4,[{file,"src/mod_http_api.erl"},{line,188}]},{mod_http_api,process,2,[{file,"src/mod_http_api.erl"},{line,141}]},{ejabberd_http,process,2,[{file,"src/ejabberd_http.erl"},{line,366}]},{ejabberd_http,process_request,1,[{file,"src/ejabberd_http.erl"},{line,488}]},{ejabberd_http,process_header,2,[{file,"src/ejabberd_http.erl"},{line,286}]}]

Start with the actual error, which is buried in the middle:
exit:{noproc,{p1_server,call,[none,{set_presence,.....
So there is a noproc error, which occurred during a call to p1_server:call, where the first argument is none. (The call would look like p1_server:call(none, {set_presence,...}).) That is, we're asking the process called none to execute the command set_presence, which fails because there is no process registered with the name none.
Now let's look at the stack trace. I added line breaks:
[{p1_server,call,3,[{file,"src/p1_server.erl"},{line,210}]},
{mod_http_api,handle2,4,[{file,"src/mod_http_api.erl"},{line,267}]},
{mod_http_api,handle,4,[{file,"src/mod_http_api.erl"},{line,228}]},
{mod_http_api,perform_call,4,[{file,"src/mod_http_api.erl"},{line,188}]},
{mod_http_api,process,2,[{file,"src/mod_http_api.erl"},{line,141}]},
{ejabberd_http,process,2,[{file,"src/ejabberd_http.erl"},{line,366}]},
{ejabberd_http,process_request,1,[{file,"src/ejabberd_http.erl"},{line,488}]},
{ejabberd_http,process_header,2,[{file,"src/ejabberd_http.erl"},{line,286}]}]
The first line is where it discovers that there is no such process, on line 210 in p1_server.erl. But we're more interested in why we're passing none as the argument in the first place. The stack trace suggests that's happening somewhere in mod_http_api.erl, but at that point I'm getting lost - it seems like it's extracting the arguments from the command, but I don't understand where the none is coming from...

How to fine-tune the level for SonarQube Gradle plugin

I'm using SonarQube plugin (version 2.6.1) for Gradle (version 4.7) and have the problem that a lot of unimportant log output is being written while running the sonar analysis on my CI server.
Is there a way to fine-tune the log level for this plugin?
I checked the documentation but the only setting related to the log output I found was the JVM argument "verbose" which I'm not using either way (I guess the default is false so this shouldn't be turned on for me).
EDIT: Here are some examples of the output I would like to get rid of:
Some huge exception stacktraces during findbugs analysis (this one is shortened, didn't want to post the whole stacktrace, it's really huge).
16:23:34.993 ERROR - Unable to create symbol table for : /opt/workspace/pipeline-1/src/main/java/com/SomeClass.java
java.lang.NullPointerException: null
at org.sonar.java.resolve.TypeAndReferenceSolver.getSymbolOfMemberSelectExpression(TypeAndReferenceSolver.java:232) ~[java-squid-2.5.1.jar:na]
at org.sonar.java.resolve.TypeAndReferenceSolver.resolveAs(TypeAndReferenceSolver.java:200) ~[java-squid-2.5.1.jar:na]
at org.sonar.java.resolve.TypeAndReferenceSolver.resolveAs(TypeAndReferenceSolver.java:182) ~[java-squid-2.5.1.jar:na]
at...
Stacktraces from PMD:
16:23:37.206 ERROR - Fail to execute PMD. Following file is ignored: /opt/workspace/pipeline-1/src/main/java/com/SomeClass.java
java.lang.RuntimeException: null
at org.objectweb.asm.MethodVisitor.visitParameter(Unknown Source) ~[asm-5.0.3.jar:5.0.3]
at org.objectweb.asm.ClassReader.b(Unknown Source) ~[asm-5.0.3.jar:5.0.3]
at org.objectweb.asm.ClassReader.accept(Unknown Source) ~[asm-5.0.3.jar:5.0.3]
at org.objectweb.asm.ClassReader.accept(Unknown Source) ~[asm-5.0.3.jar:5.0.3]
at net.sourceforge.pmd.lang.java.typeresolution.PMDASMClassLoader.getImportedClasses(PMDASMClassLoader.java:77) ~[pmd-java-5.2.1.jar:na]...
Lots of irrelevant warnings like these:
16:23:38.638 WARN - /opt/workspace/pipeline-1/src/main/java/com/SomeClass.java: Got an exception - expecting EOF, found '}'
/opt/workspace/pipeline-1/src/main/java/com/SomeClass.java:28:5: expecting RCURLY, found 'default'
16:23:38.655 WARN - /opt/workspace/pipeline-1/src/main/java/com/SomeClass.java: Got an exception - expecting EOF, found 'someVariable'
I don't know what exactly is causing these problems, but since both my app and the results of the sonar analysis are looking OK, I would like to get rid of those log outputs since they only pollute my logs on Jenkins and make them unreadable.

There's property sonar.log.level and sonar.verbose; for example:
allprojects {
sonarqube {
properties {
// property "sonar.log.level", "INFO"
property "sonar.log.level", "TRACE"
}
}
}
see the analysis parameters.

Showing parameter/arguments with Delphi Jedi JCLDebug

When an error occurs in my Delphi XE5 application, in my exception handler I am trying to get Jedi Code Library's JCLDebug.pas to show any parameters of method calls listed on the stack. I was hoping to get this information in a similar way to the way the Delphi IDE shows this information in its Call Stack debug window when you break during an exception, like this:
TestForm.DoSomething('A Test Parameter')
TestForm.TTestForm.DoSomethingClick($DF935B0)
VCL.Controls.TControl.Click
In my error handler, I get this a result similar to this with no parameter information:
[00F9701D]{Test.exe} TestForm.DoSomething$qqrx20System.UnicodeString (Line 138, "TestForm.pas")
[005E358B]{TestMonitor.exe} Vcl.Controls.TControl.Click$qqrv (Line 7340, "Vcl.Controls.pas")
I have initialised my error handling by calling during the units initialization section:
JclStackTrackingOptions:=JclStackTrackingOptions + [stStack, stRawMode];
JclStartExceptionTracking;
Then when the exception handler is invoked, I call the following to get the call stack strings:
JclLastExceptStackListToStrings(FErrors, True, False, False);
Is it possible to get the parameters the methods were called with using JCLDebug?

TFS2013 Fails to Deserialize Parameters containing "$/"

I'm using the TFS 2013 XAML Build Process Template and have some parameters that contain TFS paths (i.e. start with $/) and the build immediately fails when deserializing the parameters.
The only workaround I have is to use $\/ but this gets pushed through as it is literally, breaking things that expect $/. Is there any way I can escape the strings properly?
e.g.
Post-build script arguments: -Path1 "$/MyProject/MySolution.sln" -Path2 $/MyProject/NuGet"
MSBuild arguments: /p:PackageOutputDir="$/MyProject/NuGet"
This is causing the following type of exception when I try to run a build:
Exception Message: TF401070: Could not deserialize the JSON string provided. Details:
After parsing a value an unexpected character was encountered: \. Path 'PostActionScriptArguments', line 1, position 449. (type BuildParameterSerializationException)
Exception Stack Trace: at Microsoft.TeamFoundation.Build.Common.BuildParameter.set_Json(String value)
Inner Exception Details:
Exception Message: After parsing a value an unexpected character was encountered: \. Path 'PostActionScriptArguments', line 1, position 449. (type JsonReaderException)
Exception Stack Trace: at Newtonsoft.Json.JsonTextReader.ParsePostValue()

Resources