[
http://opencast.jira.com/browse/MH-7493?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=26190#comment-26190
]
Adam McKenzie commented on MH-7493:
-----------------------------------
Adam the agent actually ingests what it managed to record before it went down,
but doesn't try to record anything else. Is this okay behavior or should I
create an issue?
> QA RC 1.1.10282: Test Capture Errors/Recovery Scenarios
> -------------------------------------------------------
>
> Key: MH-7493
> URL: http://opencast.jira.com/browse/MH-7493
> Project: Matterhorn Project
> Issue Type: Sub-Task
> Components: Architecture & Services, QA and Release
> Affects Versions: 1.1
> Reporter: Adam Hochman
> Assignee: Adam McKenzie
> Priority: Release Blocker
> Fix For: 1.1.x (branch)
>
>
> REMEMBER TO RESOLVE AS "PASSED QA" OR "FAILED QA" at end of test. If the
> test uncovers bugs, enter these bugs as separate issues and link them to this
> resolved test. When you start a QA test, please put the issue in progress
> until it is resolved.
> Ingest Error and Recovery
> 1. Schedule a recording with the agent.
> 2. Confirm that the recording starts recording. Bring down the ingest
> server. Confirm that the recording goes into a warning state under the
> Upcoming recordings filter. Bring back up server after capture agent logs
> receive 500 errors upon ingest.
> 3. Confirm that the recording proceeds to ingest and that recording shows up
> properly in the admin UI after successful ingest.
> Unplug VGA2USB and RePlug
> 1. Schedule a recording with the agent.
> 2. While recording, unplug the vga2usb device and plug back in.
> 3. Confirm that video contained test bars during unplugged time and went back
> to normal recording after plugging the vga2usb back in. Make sure that the
> resulting audio and other video input streams are intact.
> Unplug VGA2USB before recording starts
> 1. Schedule a recording with the agent.
> 2. Confirm that test bars exist for vga recording and that the audio and
> other video streams remained entact for the entirety of the recording.
> Agent offline recovery
> 1. Schedule a recording. Unplug capture agent but make sure it doesn't have
> cached schedule of event prior to going offline.
> 2. In the middle of scheduled time, plug back in capture agent.
> 3. Confirm that the capture agent started and captured the remainder of the
> scheduled event.
> 4. After the agent uploads the recording, confirm that the recording is
> intact and processing correctly.
> Agent dies in the middle of recording
> 1. Schedule a recording.
> 2. In the middle of recording, turn off the capture agent's power.
> 3. When the capture agent is turned back on, confirm that Matterhorn is
> automatically running as a service.
> 4. Confirm that the capture agent does not try to upload the failed
> recording. (for 1.1 it should do nothing)
> 5. Confirm that related media exists in recordings cache and is processable.
> (not sure if this is an expectation but we should confirm whether this is the
> case.)
> Agent only receives current and upcoming events associated with itself
> 1. Schedule a recording.
> 2. Once the capture agent updates its schedule, look at the cached ics.
> 3. Confirm that scheduled events are limited to capture and are only current
> and upcoming.
> Agent does not update cached schedule if scheduled events remain the same
> (etags)
> 1. Schedule recordings.
> 2. Once the capture agent updates its schedule, look at the cached ics.
> 3. Make sure that there are not new or changed events. Look at capture logs
> and confirm that it doesn't poll the schedule service for new events or
> update its schedule cache.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
_______________________________________________
Matterhorn mailing list
[email protected]
http://lists.opencastproject.org/mailman/listinfo/matterhorn
To unsubscribe please email
[email protected]
_______________________________________________
