Harvest History

Owned by M
Last updated: Aug 21, 2013 by SrVersion comment

All jobs

Harvest Status in the left menu by default shows a chronological list of all jobs ever harvested with status Started in ascending order. The same does All jobs

If information is needed for jobs with other statuses (or All statuses) or other sort order, then this can be specified in the combo-boxes in the top of the page and then activated by clicking the Show button.

For each job the page shows information about the job and its status as well as information about errors (harvest errors or upload errors) and number of configurations in the job.

  • Chose JobID if you want to check details on a specific job. See Job list links.
  • Chose Harvest name if you want to check details on the history of a specific harvest definition.  See History of a harvestdefinition for details.
  • Chose Run number if you want to check details on a specific run of that harvestdefinition – note that a run can consist of multiple jobs. See Specific run listing.

In case of Harvest errors, a Restart button will appear and the operator can choose to resubmit that specific job to be harvested again. When resubmitting a failed job, the status will say 'Resubmitted' and refer to the new resubmitted job.

It is possible to go directly to the following 

The following pages with job details can be accessed from the job status row

JobID details

Clicking on a jobID on any of the harvest history pages will give you a very detailed report on the job.

This page gives all the information available about the job itself (e.g. max-bytes limit) and about the single domains included in the job.

Furthermore the page shows the complete seedlist used with the job and the complete Harvest order template as well as detailed error information in case of errors. The two latter is mainly for advanced users debugging specific crawls where things didn't go as expected.

Harvest definition job history

The history page for a harvestdefinition is the same as you can reach from the frontpage with the History buttons. This history page gives you further information for each run of the harvestdefinition: Start time, End time, number of bytes harvested and number of documents harvested. The page also show how many jobs each run consists of and how many of these that failed and eventually got resubmitted.

Specific run listing

If a link in the 'Run number' column is clicked, the jobs for the specific run of the harvests definition is listed. A scheduled harvest definition might generate a number of jobs depending on the configurations and number of domains. See Harvester design for details on how jobs are generated. 

Details on a terminated job

If you terminates a running job in the Heritrix GUI, you can view Job Details and see that the job is stopped due to "Harvesting aborted". If it is a bigger (snapshot) harvest, that includes several jobs, all the finished jobs, will appear as "Done". And only the ones that was actually stopped will appear as stopped due to "Harvesting aborted" for some domains.

All job per domain

This link is obsolete and will just advise the user to access the harvest history for a domain functionality from the domain details page.

Running jobs

The running jobs page displays details about the jobs currently being run by the harvesters. The information show here is continuously extracted from the running Heritrix instance and returned to the GUI.