We recently moved off Hudson to the latest Jenkins (1.475).
After doing so, we have noticed an increased startup time and clicking on
tabs takes longer too. Upon further investigation, these jobs typically
have a history of 100 or so builds.
Dumping the stack while waiting produces the following:
java.lang.Thread.State: RUNNABLE
at java.io.UnixFileSystem.getBooleanAttributes0(Native Method)
at
java.io.UnixFileSystem.getBooleanAttributes(UnixFileSystem.java:228)
at java.io.File.exists(File.java:733)
at hudson.model.AbstractBuild.calcChangeSet(AbstractBuild.java:828)
at hudson.model.AbstractBuild.getChangeSet(AbstractBuild.java:806)
at hudson.model.View$People.isApplicable(View.java:674)
at hudson.model.View.hasPeople(View.java:596)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
at
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
at java.lang.reflect.Method.invoke(Method.java:597)
at
org.apache.commons.jexl.util.introspection.UberspectImpl$VelMethodImpl.invoke(UberspectImpl.java:258)
at
org.apache.commons.jexl.parser.ASTMethod.execute(ASTMethod.java:104)
at
org.apache.commons.jexl.parser.ASTReference.execute(ASTReference.java:83)
at
org.apache.commons.jexl.parser.ASTReference.value(ASTReference.java:57)
at org.apache.commons.jexl.parser.ASTOrNode.value(ASTOrNode.java:61)
at
org.apache.commons.jexl.parser.ASTExpression.value(ASTExpression.java:54)
at
org.apache.commons.jexl.parser.ASTExpressionExpression.value(ASTExpressionExpression.java:56)
at
org.apache.commons.jexl.ExpressionImpl.evaluate(ExpressionImpl.java:80)
at
hudson.ExpressionFactory2$JexlExpression.evaluate(ExpressionFactory2.java:72)
...
To complicate things more, we actually run on a clustered filesystem
(gluster), this we cannot change. We have noticed internally checking the
existence of files and listing is significantly slower than a local
filesystem, but the redundancy outweighs the slight performance impact.
Before the upgrade, our Hudson was working fine on gluster. There was a
slight boot delay, but never saw the lag on clicking on tabs like Jenkins.
Did something change between distributions that you need to access the
history everytime you view, can you not cache it? Also, most of our builds
invoke shell scripts or call other java programs, there are no scm builds,
so the calcChangeSet() is always empty, is there a way of bypassing the
need of this call when scm is set to None?
