[pulseaudio-discuss] Hard CPU time limit aborts [Was: Re: memblock.c: Assertion 'pa_atomic_load(&b->pool->stat.n_imported) > 0' Aborting]

[Lennart Poettering]

On Mon, 06.04.09 10:24, Philip Walden (pwaldenlinux at comcast.net) wrote:

>> The crashed pulseaudio server does not seem to be leaving a core file  
>> (so as to generate a stack trace) that I can spot. Is there some  
>> configuration I need to set to make it do this?
>>
>> Here is the pacmd ls output:
>>
>> < ls output deleted >
> I still get pulseaudio crashes every few days or so. I have been  
> restarting the server from a terminal to hopefully get a core, but there  
> is never a core file. However, I am getting a variety of failures logged.
>
> -> E: object.c: Assertion 'pa_object_refcnt(o) > 0' failed at  
> pulsecore/object.c:53, function pa_object_ref(). Aborting.

This is fixed now in git.

> -> Soft CPU time limit exhausted, terminating.
> Hard CPU time limit exhausted, terminating forcibly.
>
> -> Soft CPU time limit exhausted, terminating.
> W: module-rescue-streams.c: Failed to move sink input 481 "Simultaneous  
> output on ALSA PCM on front:0 (Maestro3) via DMA" to  
> alsa_output.usb_device_471_110_noserial_if0_sound_card_0_alsa_playback_0.
> Hard CPU time limit exhausted, terminating forcibly.

These issues are probably the same. Do they go away if you disable
module-combine?

Lennart

-- 
Lennart Poettering                        Red Hat, Inc.
lennart [at] poettering [dot] net         ICQ# 11060553
http://0pointer.net/lennart/           GnuPG 0x1A015CC4