JIRA Crashes with "Failed to fork OS thread" Error Message
Symptoms
The JIRA instance crashes a couple of times due to OutOfMemoryError. However, a "Failed to fork OS thread" error message is shown instead of a common insufficient memory exception.
The following appears in the atlassian-jira.log
:
Cause:
java.lang.OutOfMemoryError: Failed to fork OS thread
Stack Trace: [hide]
java.lang.OutOfMemoryError: Failed to fork OS thread
at java.lang.Thread.startImpl(Native Method)
at java.lang.Thread.start(Thread.java:966)
at jcifs.util.transport.Transport.connect(Transport.java:171)
at jcifs.smb.SmbTransport.connect(SmbTransport.java:290)
at jcifs.smb.SmbSession.getChallenge(SmbSession.java:146)
...
Cause
This error occurs when you are using IBM WebSphere Application Server. It indicates that the maximum number of concurrent VM's is exceeded.
Resolution
Quoting the IBM article which covers this error: You can avoid this error by inserting a very small delay between starting successive threads (even as small as 1 millisecond). This delay can cause performance to decrease a little, but the error would disappear.