[mythtv-users] Live Tv stalls. Again

[Kevin Johnson]

Two days ago I change the disk scheduler from balanced disk i/o to
combination.
Since then I have not experienced a stall.
Probably not the real reason behind the stalls, but so far so good.


On Tue, Jul 22, 2014 at 6:51 PM, Kevin Johnson <iitywygms at gmail.com> wrote:

> Got some different errors from the frontend today.
> Not sure if they give any other clues.  This is after I updated the
> backend to the latest fixes.
>
> Jul 22 18:43:26 g430 mythfrontend.real: mythfrontend[2106]: W CoreContext
> mythplayer.cpp:3145 (PauseDecoder) Player(1): Waited 100ms for decoder to
> pause
>
> Jul 22 18:43:29 mythfrontend.real: last message repeated 9 times
>
> Jul 22 18:43:29 g430 mythfrontend.real: mythfrontend[2106]: E CoreContext
> mythplayer.cpp:3205 (DecoderEnd) Player(1): Failed to stop decoder loop.
>
> Jul 22 18:43:29 g430 mythfrontend.real: mythfrontend[2106]: I CoreContext
> mythplayer.cpp:5320 (SetDecoder) Player(1): Waited 10ms for decoder lock
>
> Jul 22 18:44:00 mythfrontend.real: last message repeated 1004 times
>
> Jul 22 18:45:01 mythfrontend.real: last message repeated 2100 times
>
> Jul 22 18:45:33 mythfrontend.real: last message repeated 1099 times
>
>
>
> On Tue, Jul 22, 2014 at 7:39 AM, Kevin Johnson <iitywygms at gmail.com>
> wrote:
>
>>
>>
>>
>> On Tue, Jul 22, 2014 at 1:46 AM, Mike Perkins <
>> mikep at randomtraveller.org.uk> wrote:
>>
>>> On 22/07/14 09:34, Jean-Yves Avenard wrote:
>>>
>>>> On 22 July 2014 20:26, Mike Perkins <mikep at randomtraveller.org.uk>
>>>> wrote:
>>>>
>>>>  I'm assuming that the error messages shouldn't be truncated like that?
>>>>> Is
>>>>> this an indication of buffer overflow / underflow?
>>>>>
>>>> those errors are typical to myth not writing to the file any longer,
>>>> or taking a long time to write.
>>>>
>>>> There are two threads at play.
>>>> The recorder, that capture the data from the capture card, and write it
>>>> to disk.
>>>>
>>>> and the reader which serve data to the mythfrontend client.
>>>> WaitForReadsAllowed errors are thrown when the reader has read
>>>> everything that there is to read and has been waiting for too long for
>>>> more data to be available.
>>>>
>>>> In the backend log, you will probably see where the bottleneck is.
>>>> Either the capture card, or the disk taking a long time to write XXX
>>>> bytes.
>>>>
>>>>  I'm talking about the log messages *themselves*, not the errors that
>>> are referred to.
>>>
>>
>>
>> They are truncated because I fail.
>>
>> Jul 21 21:46:07 mythbackend mythbackend: mythbackend[1338]: I
>> ProcessRequest mainserver.cpp:1556 (HandleAnnounce) adding: g430 as a
>> remote file transfer
>> Jul 21 21:46:52 mythbackend mythbackend: mythbackend[1338]: W
>> ProcessRequest ringbuffer.cpp:1224 (WaitForReadsAllowed)
>> RingBuf(/D2/livetv/1706_20140722044607.mpg): Taking too long to be allowed
>> to read..
>>
>> Jul 21 21:47:01  mythbackend: last message repeated 138 times
>> Jul 21 21:47:01 mythbackend mythbackend: mythbackend[1338]: E
>> ProcessRequest ringbuffer.cpp:1229 (WaitForReadsAllowed)
>> RingBuf(/D2/livetv/1706_20140722044607.mpg): Took more than 10 seconds to
>> be allowed $
>> Jul 21 21:47:04 mythbackend mythbackend: mythbackend[1338]: I
>> ProcessRequest mainserver.cpp:1436 (HandleAnnounce) MainServer::ANN Monitor
>>
>> I think it is safe to assume that since the backend has not been upgraded
>> in a month that the software is not the issue.
>> I think that upgrading the frontends did not cause this issue.  How could
>> it?
>> I can stream 3 different high def movies from the backend to 3 separate
>> computers without a hiccup. The network must be okay.
>>
>> This must be a drive starting to fail or something of that nature.
>>  Maybe the hdhr prime is causing this.  But then I would see recording
>> that failed correct?
>>
>> I would start the backend with the -v option, but it errors out with a
>> message about being a upstart job and -v is not supported.
>> So I need to read on how to start the backend with more error logging.
>> Once I have that going I can share the more detailed log files.
>>
>>
>>>
>>> --
>>>
>>> Mike Perkins
>>>
>>>
>>> _______________________________________________
>>> mythtv-users mailing list
>>> mythtv-users at mythtv.org
>>> http://www.mythtv.org/mailman/listinfo/mythtv-users
>>> http://wiki.mythtv.org/Mailing_List_etiquette
>>> MythTV Forums: https://forum.mythtv.org
>>>
>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.mythtv.org/pipermail/mythtv-users/attachments/20140724/455383bd/attachment-0001.html>