My server is giving this messages every time.
I think this is because the encoding, but is there a way to know where
is the problem ? any test I can make?
| , , |
| / \ |
| ((__-^^-,-^^-__)) Octavio Rossell Tabet |
| `-_---' `---_-' octavio@... |
| `--|o` 'o|--' http://octavio.gnu.org.ve |
| \ ` / irc.radiognu.org #gnu |
| .: :. |
| :o_o: Huella: FC69 551B ECB9 62B0 D992 |
| "-" BE57 B551 2497 C78B 870A |
From: David Baelde <david.baelde@gm...> - 2010-08-24 00:48:41
I have just started a FAQ, including this question -- I forgot to
check whether my previous answers on this list included more details
or clearer explanations. Here is the FAQ entry:
This error means that a clock is getting late in liquidsoap. This can
be caused by an overloaded CPU, if your script is doing too much
encoding or processing: in that case, you should reduce the load on
your machine or simplify your liquidsoap script. The latency may also
be caused by some lag, for example a network lag will cause the
icecast output to hang, making the clock late.
The first kind of latency is problematic because it tends to
accumulate, eventually leading to the restarting of outputs: Too much
latency! Resetting active source...
The second kind of latency can often be ignored: if you are streaming
to an icecast server, there are several buffers between you and your
listeners which make this problem invisible to them. But in more
realtime applications, even small lags will result in glitches.
In some situations, it is possible to isolate some parts of a script
from the latency caused by other parts. For example, it is possible to
produce a clean script and back it up into a file, independently of
its output to icecast (which again is sensitive to network lags). For
more details on those techniques, read about clocks.
Hope that helps,