[mythtv-users] crash after ringbuff limit

[Oliver Wardell]

I seem to be having a problem where once the ringbuffer ,ax size has been reached (5GB in this case) I get the error:

ERROR: file I/O problem in 'safe_read()'
: Bad address

repeated in the backend log and the frontend log shows:

Waited 2 seconds for data to become available, waiting again...

repeatedly.

Has this been seen before? I seem to have had it in ever version of myth I've run.

The specs are:

Gentoo running with the ac kernel. myth 0.13. PVR 250. backend runs as root and writes the ringbuffer to a large hard disk (7200 rpm, 8MB cache) sat on a serial ide (asus delux mobo). The front end runs as user myth. Does this user require write access to the ringbuf file??? It currently only has read as its is written by root.

The disk access is fine:


TangleVision-live root # hdparm -tT /dev/hdg

/dev/hdg:
 Timing buffer-cache reads:   1772 MB in  2.00 seconds = 886.00 MB/sec
 Timing buffered disk reads:  166 MB in  3.03 seconds =  54.79 MB/sec


Any help gratefully received,

Cheers
Oliver
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mythtv.org/pipermail/mythtv-users/attachments/20040108/da3a5fac/attachment.html