[Haifux] Re: New haifux site

2003-12-25 Thread Shlomi Fish
On Wed, 24 Dec 2003, Orna Agmon wrote: On Wed, 24 Dec 2003, Ron Artstein wrote: I am proud to present the new haifux site, located as (almost) always at: http://www.haifux.org Three comments: 1. The link to the mailing list archives stops in early April 2003; the traffic

[Haifux] Immortal artsd process

2003-12-25 Thread Eli Billauer
Hello list. This question has been asked elsewhere, but I haven't seen anyone answering it. I have occasional problems with artsd (version 1.0.0, as in RH7.3), which stops functioning (i.e. no sound). It also takes 98% of my CPU, and a quick look at its /proc entry shows that it has a modest

Re: [Haifux] Immortal artsd process

2003-12-25 Thread Muli Ben-Yehuda
On Thu, Dec 25, 2003 at 06:06:24PM +0200, Eli Billauer wrote: The issue is that I can't kill the process. The process number is 11005, I go kill -9 11005 as root, and nothing happens (no, there is no process number confusion here). This looks like a kernel problem to me (I'm using 2.4.21).

Re: [Haifux] Immortal artsd process

2003-12-25 Thread Muli Ben-Yehuda
On Thu, Dec 25, 2003 at 06:48:05PM +0200, Eli Billauer wrote: Muli Ben-Yehuda wrote: Usually, you'll see a processes in an uninterruptible sleep represented as 'D' in ps. Since you say they are 'R', please do magic-sysrq-t to see where in the kernel they are. For those of you who have no

Re: [Haifux] Immortal artsd process

2003-12-25 Thread Eli Billauer
Muli Ben-Yehuda wrote: snip Not much. It's a backtrace, but it needs to be resolved. Please use either 'addr2line' or 'ksymoops' to translate the address (c0109ea8) to a function in the kernel. You'll need the vmlinux for your kernel to do that. This is what ksymoops translated for me

Re: [Haifux] Immortal artsd process

2003-12-25 Thread Muli Ben-Yehuda
On Thu, Dec 25, 2003 at 08:53:36PM +0200, Eli Billauer wrote: Trace; c0109ea8 do_IRQ+68/b0 This is probably irrelevant. In Linux, unless configured differently, interupt handlers run on the stack of the process that was executing at the time we took the interrupt. Trace; d0ee7f7f