-
Bug
-
Resolution: Fixed
-
Minor
-
None
-
Jenkins 2.19.1
Timestamper 1.8.7 (but observed in older versions as well)
When a single pipeline step generates a lot of logging all at once, the log on disk will repeat itself without end, eventually running the host out of disk space.
This persists between Jenkins restarts and can only be stopped by stopping Jenkins, deleting the $JENKINS_HOME/jobs/[job]/build/[build id] directory, and starting Jenkins again.
STR:
1. Create a new pipeline job with the following pipeline script definition:
stage ('Boom') {
node {
timestamps
}
}
2. Run the job.
Expected result:
Job terminates eventually.
Actual result:
Job does not terminate, and fills the log continuously.
[JENKINS-40762] Logs can grow without bound when using timestamper plugin
Edwin Flores
created issue -
Edwin Flores
made changes -
| Description |
Original:
When a single pipeline step generates a lot of logging all at once, the log on disk will repeat itself without end, eventually running the host out of disk space. This persists between Jenkins restarts and can only be stopped by stopping Jenkins, deleting the $JENKINS_HOME/jobs/[job]/build/[build id] directory, and starting Jenkins again. STR: 1. Create a new pipeline job with the following pipeline script definition: {{stage ('Boom') { node { timestamps { sh "seq 1 100000" } } }}} 2. Run the job. Expected result: Job terminates eventually. Actual result: Job does not terminate, and fills the log continuously. |
New:
When a single pipeline step generates a lot of logging all at once, the log on disk will repeat itself without end, eventually running the host out of disk space. This persists between Jenkins restarts and can only be stopped by stopping Jenkins, deleting the $JENKINS_HOME/jobs/[job]/build/[build id] directory, and starting Jenkins again. STR: 1. Create a new pipeline job with the following pipeline script definition: {{ stage ('Boom') { node { timestamps { sh "seq 1 100000" } } } }} 2. Run the job. Expected result: Job terminates eventually. Actual result: Job does not terminate, and fills the log continuously. |
Edwin Flores
made changes -
| Description |
Original:
When a single pipeline step generates a lot of logging all at once, the log on disk will repeat itself without end, eventually running the host out of disk space. This persists between Jenkins restarts and can only be stopped by stopping Jenkins, deleting the $JENKINS_HOME/jobs/[job]/build/[build id] directory, and starting Jenkins again. STR: 1. Create a new pipeline job with the following pipeline script definition: {{ stage ('Boom') { node { timestamps { sh "seq 1 100000" } } } }} 2. Run the job. Expected result: Job terminates eventually. Actual result: Job does not terminate, and fills the log continuously. |
New:
When a single pipeline step generates a lot of logging all at once, the log on disk will repeat itself without end, eventually running the host out of disk space. This persists between Jenkins restarts and can only be stopped by stopping Jenkins, deleting the $JENKINS_HOME/jobs/[job]/build/[build id] directory, and starting Jenkins again. STR: 1. Create a new pipeline job with the following pipeline script definition: stage ('Boom') { node { timestamps { sh "seq 1 100000" } } } 2. Run the job. Expected result: Job terminates eventually. Actual result: Job does not terminate, and fills the log continuously. |
Steven G Brown
made changes -
| Resolution | New: Fixed [ 1 ] | |
| Status | Original: Open [ 1 ] | New: Resolved [ 5 ] |
