-----Original Message----- From: vdr-bounces@linuxtv.org [mailto:vdr-bounces@linuxtv.org] On Behalf Of Joerg Knitter Sent: Wednesday, February 16, 2005 2:40 AM To: Klaus Schmidinger's VDR Subject: Re: [vdr] vdr-1.3.21: video data stream broken
...
Installed VDR 1.3.21 yesterday with CVS drivers from 25.01.2005, and while I did not see any problems with VDR 1.3.19 I previously had installed, I immediately got VDSB messages with VDR 1.3.21 when I recorded with the FF card while I was zapping with the Nova-S (because the FF was "occupied").
Just sharing my experience, I'm using 1.3.21 since Monday, using Nexus with SS2 2.6c and I was zapping whole evening while FF card was occupied and could not get VDSB to come up. I used enAIO, jumpplay and several other minor patches. If I occupy budget card and start zapping with FF card, I'll get VDSB very quick. Budget card seems to take longer before it locks on signal, takes few secs, FF card gets lock nearly instantly. If FF card is on same channel where budget card starts the timer, and later on FF card starts recording on another channel, I'll get VDSB and after restart both recordings would complete without a problem. I did not notice any changes if I would reload drivers or not after VDSB, or even after reboot, system behaves the same. Last night, system went through 7 timers without VDSB, there were two times where both cards were occupied.
Vanilla kernel 2.6.10 w/ CVS DVB drivers from last week. Cards have their own IRQs, only NIC and IDE controller share same IRQ, everything else has its own IRQ in the system.
Vladimir Shved wrote:
behaves the same. Last night, system went through 7 timers without VDSB, there were two times where both cards were occupied.
Same here :-( All recordings I did yesterday were okay, even recordings on both devices at the same time. Need to test some more to get it reproduceable.
With kind regards
Joerg
Joerg Knitter wrote:
Vladimir Shved wrote:
behaves the same. Last night, system went through 7 timers without VDSB, there were two times where both cards were occupied.
Same here :-( All recordings I did yesterday were okay, even recordings on both devices at the same time. Need to test some more to get it reproduceable.
Are you absolutely positively sure this is really a new problem?
Just as a reminder, the test case for the old "video data stream broken" problem is:
* You have to have two (or more) cards, the second one being a budget card. * Kernel 2.6. * EPG scan must be enabled. * After the EPG scan has first started, start a recording on the second card.
Fails here every time.
Carsten.
Carsten Koch schrieb:
Joerg Knitter wrote:
Vladimir Shved wrote:
behaves the same. Last night, system went through 7 timers without VDSB, there were two times where both cards were occupied.
Same here :-( All recordings I did yesterday were okay, even recordings on both devices at the same time. Need to test some more to get it reproduceable.
Are you absolutely positively sure this is really a new problem?
Yes, because it was said to be fixed beginning of this year, and read the positive feedback at vdrportal. Additionally, I did not have any video data stream error as long I was using VDR 1.3.19 - but I used this only for a few weeks.
With kind regards
Joerg
one Nexus-s, one Nova. kernel 2.4 EPG-Scan is disabled.
crash, when doing 2 recordings while zapping.
--Am Donnerstag, 17. Februar 2005 12:56 +0100 schrieb Carsten Koch Carsten.Koch@icem.com:
Are you absolutely positively sure this is really a new problem?
Just as a reminder, the test case for the old "video data stream broken" problem is:
- You have to have two (or more) cards, the second one being a budget
card.
- Kernel 2.6.
- EPG scan must be enabled.
- After the EPG scan has first started, start a recording on the second
card.
Fails here every time.
Carsten.
vdr mailing list vdr@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/vdr
-
Carsten Koch -
Joerg Knitter -
sn123py vdr user -
Vladimir Shved