Skip to content

Fix start-workload.sh#62

Closed
sunchao wants to merge 1 commit into
linkedin:masterfrom
sunchao:fix-workload
Closed

Fix start-workload.sh#62
sunchao wants to merge 1 commit into
linkedin:masterfrom
sunchao:fix-workload

Conversation

@sunchao

@sunchao sunchao commented Aug 28, 2018

Copy link
Copy Markdown
Contributor

This fixes two issues encountered when running start-workload.sh:

  1. The README description is incorrect. It should use = for
    config key & value for auditreplay configs.
  2. The mapper_class_name value doesn't work - it should look under
    com.linkedin.dynamometer.workloadgenerator.audit instead of
    com.linkedin.dynamometer.workloadgenerator

Fixes #60 .

This fixes two issues encountered when running start-workload.sh:

1. The README description is incorrect. It should use `=` for
   config key & value for auditreplay configs.
2. The mapper_class_name value doesn't work - it should look under
   com.linkedin.dynamometer.workloadgenerator.audit instead of
   com.linkedin.dynamometer.workloadgenerator
private Class<? extends WorkloadMapper> getMapperClass(String className) throws ClassNotFoundException {
if (!className.contains(".")) {
className = WorkloadDriver.class.getPackage().getName() + "." + className;
className = WorkloadDriver.class.getPackage().getName() + ".audit." + className;

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This will fix things for the AuditReplayMapper, but break things for CreateFileMapper... I created the audit package to group together related classes, but didn't realize I broke the assumption that all mapper classes would be in the same package. I was trying to be clever here, but maybe it would be better to just maintain a static mapping of short name -> class.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Error when running start-workload.sh

2 participants