[vdr] Re: Buffer overflows

[Sascha Volkenandt <lists@magoa.net>]

At Donnerstag, 25. September 2003 22:14 Jan Ekholm wrote:
> Anyone know if increasing the various buffers within VDR is enough to get
> rid of the buffer overflows? Is it a true solution or just band aid?
> There was a mail a while ago that mentioned values that could be used and
> where, but I've not tried that yet. But tonight I missed a South Park
> episode due to an overflow, so I could give the idea a shot. :)

Usually a buffer overflow only occurs if, and only if a receiver cannot 
process any more data (i.e. recorder: the disk can't accept data, network: 
the client doesn't read data fast enough, ...). So only buffer overflows in 
the recorder are fatal (in your case). 

If a receiver can't accept any more data, that can have various cases... 
When the data is erroneous, some players don't read any further, but don't 
close the stream either, so VDR can't send out any more data. That means 
buffers will overflow (since they are not emptied anymore). 

Since VDR 1.1.something (don't remember currently), VDR itself should not be 
affected by buffer overflows anymore (aside from the syslog messages), 
since overflowing data is ignored "noisily".

> Can someone point me towards some RTFM material that would explain what
> causes the buffer overflows, and why it is so fatal?

Apart from some posts on the list, VDR's source code is the only manual 
available on that :). Like I said, it is (or should be) fatal only if you 
are recording. The only cases I could imagine for that would be a) your 
harddisks couldn't write fast enough (maybe a background-process started?) 
or the delivered material was erroneous (bad weather?)

Greetings,
Sascha




-- 
Info:
To unsubscribe send a mail to ecartis@linuxtv.org with "unsubscribe vdr" as subject.