Skip to main content

TSB not started intermittently.

7 replies [Last post]
sivakumaran
Offline
Joined: 2006-05-05
Points: 0

In our testing in RI simulator, intermittently the buffering is not started as expected. We are able to see the below logs continuously. Please find the attached logs.

Line 31888: 20120410 03:56:06.080 INFO RI.Pipeline.TSB- tsbStateStarted -- Waiting for TSB to start operation
Line 31889: 20120410 03:56:06.330 INFO RI.Pipeline.TSB- tsbStateStarted -- Waiting for TSB to start operation
Line 31890: 20120410 03:56:06.580 INFO RI.Pipeline.TSB- tsbStateStarted -- Waiting for TSB to start operation
Line 31891: 20120410 03:56:06.830 INFO RI.Pipeline.TSB- tsbStateStarted -- Waiting for TSB to start operation
Line 31892: 20120410 03:56:07.080 INFO RI.Pipeline.TSB- tsbStateStarted -- Waiting for TSB to start operation
Line 31893: 20120410 03:56:07.330 INFO RI.Pipeline.TSB- tsbStateStarted -- Waiting for TSB to start operation
Line 31894: 20120410 03:56:07.580 INFO RI.Pipeline.TSB- tsbStateStarted -- Waiting for TSB to start operation

Once the above problem is occured, most of the time, the buffering never started in platform and few cases, after some time, the buffering started by platform but the respective TSW is no more active in RI as the initiator[testcase] already requested to stop the buffering but the RI did not honor the request as the "buffering" flag in TSBImpl is false at that time and throws IllegalStateException on calling stopBuffering(). So the native stop is not called which leads to have active buffering in platform later. This restricts to start new buffering on the same service which throws IllegalStateException on calling startBuffering().

Please comment and let me know if anybody faced this issue earlier.

AttachmentSize
RILog-tipri-tsbfailure.zip1.38 MB

Reply viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
smaynard
Offline
Joined: 2009-01-27
Points: 0

We experienced this problem long ago with older versions of code. It was thought to have been fixed as we have not reproduced this in many months.

What version of code are you running and can you include any other test set-up specific information that may help us understand how you are reproducing this?

Thanks,
Steve

sivakumaran
Offline
Joined: 2006-05-05
Points: 0

We are using RI 1.2 revision 31517. This issue is randomly happening on the testcases related to buffering and recording. In our testing with TDK, we are running the good and all list testcases and mostly the issue got observed in TC2118, TC2119, TC0633 testcases. In most of the cases, the TSB starts after some time in native and further attempt from RI to start buffering fails, because of this, lot of other testcases are failing.

RI.Stack- 9689421 [ThreadPool-System-13] INFO recording.RecordingImpl - RI 0xdf234fd2: attachForBufferingAndRecording: Unexpected Exception
java.lang.IllegalStateException: Error starting TSB (MPE Error 1)
at org.cablelabs.impl.manager.timeshift.TimeShiftWindow.startBuffering(TimeShiftWindow.java:1714)
at org.cablelabs.impl.manager.timeshift.TimeShiftWindow.updateForAddedUses(TimeShiftWindow.java:1179)
at org.cablelabs.impl.manager.timeshift.TimeShiftWindow$TimeShiftWindowClientImpl.attachFor(TimeShiftWindow.java:4470)
at org.cablelabs.impl.manager.recording.RecordingImpl.attachForBufferingAndRecording(RecordingImpl.java:657)
at org.cablelabs.impl.manager.recording.RecordingImpl.access$500(RecordingImpl.java:166)
at org.cablelabs.impl.manager.recording.RecordingImpl$IStateWaitTuneSuccess.processTswEventReadyToBuffer(RecordingImpl.java:4841)
at org.cablelabs.impl.manager.recording.RecordingImpl$IState.handleTswStateChange(RecordingImpl.java:4458)
at org.cablelabs.impl.manager.recording.RecordingImpl.tswStateChanged(RecordingImpl.java:5252)
at org.cablelabs.impl.manager.timeshift.TimeShiftWindow$1.run(TimeShiftWindow.java:982)

ssathish
Offline
Joined: 2011-05-26
Points: 0

Hi Steve,

Any updates on this issue?

smaynard
Offline
Joined: 2009-01-27
Points: 0

We have not been able to reproduce this. Are you still seeing the same problem on the latest trunk? If so, please file an issue with logs (preferably set to DEBUG level) and any additional data that may help us in reproducing the problem.

ssathish
Offline
Joined: 2011-05-26
Points: 0

The issue is rarely reproducible. In RI 122 version, the same issue got reproduced on 03Jul2012.
As it is rarely reproducible, couldn't able to get DEBUG level enabled logs. I am trying to get it.

Meanwhile, please find the attached log --WaitingTSB_122_03Jul2012.txt--

smaynard
Offline
Joined: 2009-01-27
Points: 0

I think an issue should still be filed with all the logs captured so far attached. That way if it continues to be elusive we can slowly build up an information base to draw from for diagnosis.

ssathish
Offline
Joined: 2011-05-26
Points: 0

Filed an issue tracker: OCAP_RI-655

Thanks