Process that wouldn't die. (Summary)

From: hamp@bd3100.idi.battelle.org
Date: Tue Apr 06 1993 - 15:16:56 CDT


I wrote:

> On SUN OS 4.1.3, our database daemon gets stuck in this state:
> (ps -axl)
>
> 24088001 300 179 1 0 0 -13944 84 clk_intr S < ? 0:01 dmka1_l1_k01
>
> It is not killable. Can anyone fill me in as to what that state is?
> (Or point me in the right documentation route?)

---
I got two responses.

stern@sunne.East.Sun.com (Hal Stern - NE Area Systems Engineer)
said:

i'd suggest getting a kernel stack trace of the process
to see where it's stuck. the attached document has
details on how to do this.

(He included a post script file titled "Building and Debugging SunOS
Kernels")

ups!kalli!kevin@fourx.Aus.Sun.com (Kevin Sheehan {Consulting Poster Child})
said:

It looks like the process is sleeping in the kernel waiting for a clock
tick or timeout of some kind. You can use adb -k /vmunix /dev/mem and
the $<traceall macro to see the call graph and figure out more...

---

Thanks for the responses. This should give me enough info to go on. Of course, since I posted this, we have not had this hang happen again!

Don



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:07:41 CDT