Subject: Re: [ecasound] eci problem
From: Kai Vehmanen (k_AT_eca.cx)
Date: Thu May 15 2003 - 23:55:56 EEST
On 14 May 2003, Adam Linson wrote:
> (ecasoundc_sa) Error='read() error', cmd='c-select reader1'
> last_error=''
> cmd_cnt=2532 last_cnt=2531.
This is either caused by a) bug in libecasoundc, or b) bug in ecasound
that caused it to crash or fail in some other way. As the first step you
could use 'pstree' to see if the child ecasound process disappers when
this error occurs.
> the precise step during which the error is encountered varies (i.e., it
> is not always "c-select", but any other number of commands). when this
> problem happens, it has usually been running for at least 10 minutes, at
> most 45 minutes (i would like it to run indefinitely, or at least for 2
> hours).
Hmm, try uncommenting the line
/* #define ECI_ENABLE_DEBUG */
in 'ecasound/libecasoundc/ecasoundc_sa.c', and then recompile your app.
You should see more debugging info. Also, you can increase the engine
verbosity by issuing "debug 511" ECI command at the start of your program.
I'd be interested in the logs there produce before and after the bug
occurs.
> i am running RH8 with low-latency patches and ecasound 2.2.1, compiled
> with large-file-support. it generally runs without any glitch
> whatsoever. the above problem happens whether X is running or not, and i
> always run it as root.
Yup, the ECI functionality should work just fine without any low-latency
patches (i.e. it's not timing sensitive).
-- http://www.eca.cx Audio software for Linux!
This archive was generated by hypermail 2b28 : Thu May 15 2003 - 23:57:27 EEST