| View previous topic :: View next topic |
| Author |
Message |
Jogie214
Tux's lil' helper
Joined: 15 Aug 2004
Posts: 144
Location: Cologne / Germany
|
 Posted: Sun Mar 11, 2007 12:14 pm Post subject: Can't kill rsync |
 |
|
Hello!
I have 10 rsync processes, and I can't kill anyone of it, I tried kill -9, -15 (to be honest I tried every signal form -1 to -15), is there anything else I can do? I know, I can reboot, but then all my friends who tell me ubuntu is more stable would be right, and I don't like that idea....  .
Thanks
Seb
_________________
Desktop: Core 2 Duo E6600 / Asus P5B / 4GB DDR2-800 / Ati HD-4850/ Sidux amd64
Server : Phenom2 905e / Asus M4A785TD-V / 8GB DDR3-1333 / Geforce 8600GT / Gentoo amd64
Laptop : Core 2 Duo 5500 / HP nx7300 / 4 GB-RAM / Intel / Gentoo amd64 |
|
| Back to top |
|
 |
iphands
n00b
Joined: 01 Oct 2005
Posts: 61
|
| Posted: Sun Mar 11, 2007 4:31 pm Post subject: |
|
|
what about killall
should kill all processes with that name...
Why 9? If you don't mind my asking. |
|
| Back to top |
|
|
Jogie214
Tux's lil' helper
Joined: 15 Aug 2004
Posts: 144
Location: Cologne / Germany
|
| Posted: Sun Mar 11, 2007 10:11 pm Post subject: |
|
|
Hi!
Sorry, I forgot to mention it, that I also tried killall and kill .. $(pidof rsync).
To be honest I'm not able to explain why -9, I always used -9 or -15....
BTW, the rsync processes still occupy memory, they are no zombie processes.
Seb
_________________
Desktop: Core 2 Duo E6600 / Asus P5B / 4GB DDR2-800 / Ati HD-4850/ Sidux amd64
Server : Phenom2 905e / Asus M4A785TD-V / 8GB DDR3-1333 / Geforce 8600GT / Gentoo amd64
Laptop : Core 2 Duo 5500 / HP nx7300 / 4 GB-RAM / Intel / Gentoo amd64 |
|
| Back to top |
|
|
Hu
Watchman
Joined: 06 Mar 2007
Posts: 7607
|
| Posted: Sun Mar 11, 2007 11:34 pm Post subject: |
|
|
| Please post the output of ps auxwwwww | grep rsync and for a in $(pgrep rsync); do /usr/sbin/lsof -p $a; done. ps is part of sys-process/procps; lsof is from sys-process/lsof. I suspect the rsync is in a device wait state, since that is the only way to survive a SIGKILL. Note that -9 is short for -KILL, which sends SIGKILL, a non-blockable fatal signal. Processes which receive a SIGKILL are immediately terminated and cannot execute any cleanup code. However, if a process is in a device wait state, the kernel may defer handling the kill until the device wait is completed. |
|
| Back to top |
|
|
iphands
n00b
Joined: 01 Oct 2005
Posts: 61
|
| Posted: Sun Mar 11, 2007 11:36 pm Post subject: |
|
|
| Jogie214 wrote: | Hi!
Sorry, I forgot to mention it, that I also tried killall and kill .. $(pidof rsync).
To be honest I'm not able to explain why -9, I always used -9 or -15....
BTW, the rsync processes still occupy memory, they are no zombie processes.
Seb |
Weird...
Oh no -9 is kill (really kill) without waiting for the process to close up or do any clean up. Its kinda like pulling the power on a process.
I meant why 10 (the number of rsync processes) how did they get there. I am at a loss as to why they aren't getting killed properly. |
|
| Back to top |
|
|
Captain Newbie
Apprentice
Joined: 22 Dec 2006
Posts: 174
Location: Socal
|
| Posted: Mon Mar 12, 2007 12:00 am Post subject: |
|
|
| iphands wrote: | | Jogie214 wrote: | Hi!
Sorry, I forgot to mention it, that I also tried killall and kill .. $(pidof rsync).
To be honest I'm not able to explain why -9, I always used -9 or -15....
BTW, the rsync processes still occupy memory, they are no zombie processes.
Seb |
Weird...
Oh no -9 is kill (really kill) without waiting for the process to close up or do any clean up. Its kinda like pulling the power on a process.
I meant why 10 (the number of rsync processes) how did they get there. I am at a loss as to why they aren't getting killed properly. |
KILL's not supposed to be caught or ignored like that (obviously)...let's see that output Hu asked for.
_________________
jmp resume_userspace
"As much as it pains me, we hope that developers know what they're doing." - wolf31o2 |
|
| Back to top |
|
|
Jogie214
Tux's lil' helper
Joined: 15 Aug 2004
Posts: 144
Location: Cologne / Germany
|
| Posted: Mon Mar 12, 2007 11:30 am Post subject: |
|
|
Thanks everybody!
But I had to reboot in the mean time, because I needed to emerge iverilog (I need it for my diploma work), and I wasn't able to emerge anything, apparently my portage got corrupted (only God knows how), I wasn't even able to delete or move portage until I rebooted, know I'm able to sync again...
Checkfs hasn't found any errors on /usr/portage/ during the reboot, so some of the portage files must have been looked.
BTW, I erased my whole /usr/portage (previously xfs) and use ext4dev from now one, ext4 is much faster with this huge number of small files.
Thanks again
Seb
P.S.: Sorry for my bad english
_________________
Desktop: Core 2 Duo E6600 / Asus P5B / 4GB DDR2-800 / Ati HD-4850/ Sidux amd64
Server : Phenom2 905e / Asus M4A785TD-V / 8GB DDR3-1333 / Geforce 8600GT / Gentoo amd64
Laptop : Core 2 Duo 5500 / HP nx7300 / 4 GB-RAM / Intel / Gentoo amd64 |
|
| Back to top |
|
|
Stephen_Cataldo
n00b
Joined: 08 May 2012
Posts: 1
|
| Posted: Tue May 08, 2012 10:58 pm Post subject: piped through ssh? |
|
|
I had processes like these, with rsync running over ssh:
m2tech 25467 0.4 -0.2 29876 3156 ?? S Sat12AM 19:30.32 ssh -l svpadmin example.org rsync --server ...
m2tech 524 0.4 -0.1 29544 1756 ?? S Wed12AM 51:46.28 rsync -verbose -iavHz --progress --numeric-ids -e ssh ...
killall -9 rsync
did NOT work for me, but this did:
killall -9 ssh |
|
| Back to top |
|
|
|
Portage & Programming
