We help IT Professionals succeed at work.

How to kill a process...

ruamire
ruamire asked
on
Medium Priority
2,266 Views
Last Modified: 2013-12-06
Hi,

I have 3 processes that just won't die, even with kill -9. I'm not aware of any other more severe way to kill a process.

This is on HP-UX.

If you can advise, please do!!

Thanks,
Derek.
Comment
Watch Question

Guy Hengel [angelIII / a3]Billing Engineer
CERTIFIED EXPERT
Most Valuable Expert 2014
Top Expert 2009
Commented:
which processes? can you show the ps -eaf  for the relevant processes?

Not the solution you were looking for? Getting a personalized solution is easy.

Ask the Experts

Author

Commented:
BTW : I am aware of a more severe way, a reboot, but I'm not that keen on that solution.

Author

Commented:
Here's the output you requested....

  bt21cd 15352     1  0 13:41:19 ?         2:09 /net/burns/export/home2/bt21cd/ac/f3.6/b
  bt21cd 15029     1  0 13:30:41 ?         2:53 /net/burns/export/home2/bt21cd/ac/f3.6/b
  bt21cd 16373     1  0 14:35:32 ?        76:15 /net/burns/export/home2/bt21cd/ac/f3.6/b

Author

Commented:
I've ran kill -9 15352 15029 16373
to no avail...

Author

Commented:
These are not zombie processes, they are holding ports which means I cannot start another instance of my process.
CERTIFIED EXPERT
Top Expert 2006
Commented:
have u tried the fuser -f to find  which files or directories within a filesystem are currently in use and to identify the processes and users that are using them
after that you can use fuser -k
if they have guis use xkill

or try user logoff and relogin

when they hold ports it might be worse a network restart

rcnetwork restart
Commented:
If you have access to a serial console, try going into Single User mode - that'll kill all user proc's and still avoid going <completely> down.  init S
if kill -9 does not stop the process, then your user have not proper permissions, or the process is disk waiting (seems to be started from NFS), or it's a zombie
Top Expert 2015
Commented:
It lost the file handle on NFS down below, has pending writes , and will not die. Data has been damaged.
System reboot is 100% working fix. But try single mode first.

To avoid:
Do not mount NFS over firewalls and use TCP instead of UDP for involved mount
Commented:
Hey - if you're using nfs, check for lockd, statd daemons.  They've been known to flake out (at least on Solaris) and restarting those daemons could fix the problem.
On solaris REBOOT

Author

Commented:
Unfortunately none of these options worked. I had to reboot HP-UX.
Top Expert 2015
Commented:
Did you take measures to avoid problem in future?
> Unfortunately none of these options worked. I had to reboot HP-UX.
could you please explain?

Are you saying that user root cannot kill these processes, they are no zombies, and they are not disk waiting?
unbelievable ...
Top Expert 2015

Commented:
Asker took Window approach - reboot when in doubt .... No need to solve problem.
Windows approach is reinstall from scratch, isn't it?
SCNR :-))

Author

Commented:
I'm going to close this off, hopefully won't be in this situation again anytime soon.

Points for all cause I can't say any particular solution worked but thanks for the comments.
Access more of Experts Exchange with a free account
Thanks for using Experts Exchange.

Create a free account to continue.

Limited access with a free account allows you to:

  • View three pieces of content (articles, solutions, posts, and videos)
  • Ask the experts questions (counted toward content limit)
  • Customize your dashboard and profile

*This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

OR

Please enter a first name

Please enter a last name

8+ characters (letters, numbers, and a symbol)

By clicking, you agree to the Terms of Use and Privacy Policy.