Move logging from BaseScript into JobRunner

[alehaa]

NetBox version

v4.1-beta1

Feature type

Change to existing functionality

Proposed functionality

• Move the logging methods (i.e. self.log_info()) from extras.scripts.BaseScript into netbox.jobs.JobRunner.
• Move job log views from script to general job details.

Use case

With the JobRunner framework, plugins have the ability to define jobs in a standard way. However, communication to the user is still limited to free-form data and exception logging. Adding logging to general jobs would allow plugins to provide additional information about what's been done during job execution.

Database changes

None

External dependencies

None.

If accepted, this could make #15983 obsolete. As all jobs would have logs, no dedicated link to script logs is required anymore.

[alehaa]

I volunteer to work on this myself.

[jeremystretch]

I'm not sure I understand the intent here. Logging methods like log_info() are intended to be called by the script itself within its run() method to log output from its operations. @alehaa could you elaborate a bit on the specific change(s) being proposed?

[alehaa]

My intention when introducing the JobRunner framework was to provide a standard way to handle jobs. Previously there were some different implementations in NetBox and also some plugins. With NetBox v4.1 it is now possible to access this standard interface and e.g. show which background jobs are running along with other jobs like scripts or data source syncs.

With this FR I'd like to introduce the next evolution of the framework and give it the ability to communicate information to users and administrators in a standard way. Currently, scripts, the blueprint of the JobRunner framework, have a standardized way to communicate messages using logging methods. Now I'd like to make logging available to the JobRunner class as well. This would allow background jobs and housekeeping tasks to document what they've done, e.g. which VMs were synchronized or skipped, just like scripts do.

To avoid reinventing the wheel, I suggest moving the existing methods from the BaseScript class to JobRunner or a common base class (to be introduced). The JobRunner class has been designed to keep extensions like this in mind, so one would call log_info() from run() like in scripts. The UI log views also need to be moved from scripts to jobs.

[alehaa]

@jeremystretch do you need some additional information? It's unclear to me whether this issue is in the state needs owner or revisions needed.

[alehaa]

I am still happy to work on this feature. Since there won't be any model changes, there shouldn't be any significant changes to the existing functionality, just making it available to other NetBox internal and external code.

[fdietzsch]

I'd love to see this feature. I am currently using a @system_job, a feature of the recent NetBox 4.2 release. Currently I don't know how to communicate details about the job execution to the user. For example I'd like to see on which device objects the current execution operates.

Any current workaround is highligh appreciated.

This is my current system_job

@system_job(interval=JobIntervalChoices.INTERVAL_MINUTELY)
class EnsureMaintenanceStatusJob(JobRunner):
    class Meta:
        name = "Check device status"
        description = "Set 'Maintenance' status for Devices if current time is within the maintenance window."


    def run(self, *args, **kwargs):
        now = timezone.now()

        # obtained via Django field lookups, e.g. https://docs.djangoproject.com/en/3.2/ref/models/querysets/#gte
        active_mw = DeviceMaintenance.objects.filter(start_time__lte=now, end_time__gte=now)
        devices_in_maintenance = active_mw.values_list('device_id', flat=True)


        # !!! This is the highly appreciated part, e.g.
        # self.log_info("Number of devices in maintenance: ", len(active_mw))
        # !!! 

        # filter for all devices that have a maintenance window and set their status accordingly
        Device.objects.filter(pk__in=devices_in_maintenance).update(status='maintenance')

        # search all devices that have maintenance status but are no longer within a maintenance window
        revert_qs = Device.objects.filter(status='maintenance').exclude(pk__in=devices_in_maintenance)
        revert_qs.update(status='staged')

[alehaa]

@jeremystretch I think this issue is stuck. Can you give any advice on how to proceed? I'm happy to provide a PR for this on the feature branch for the upcoming 4.3 release.

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. NetBox is governed by a small group of core maintainers which means not all opened issues may receive direct feedback. Do not attempt to circumvent this process by "bumping" the issue; doing so will result in its immediate closure and you may be barred from participating in any future discussions. Please see our contributing guide.

[github-actions[bot]]

This issue has been automatically closed due to lack of activity. In an effort to reduce noise, please do not comment any further. Note that the core maintainers may elect to reopen this issue at a later date if deemed necessary.