ps -ef | grep results; what does the ? mean?

[stevenriz]

Hello, this NT mount recently became unresponsive and I don't know why or what to do short of restarting the server altogether. DS1 is a Windows server with a windows Share we have mounted onto s1; s1 being the linux machine we run the mount command onto.... Well during a script run, DS1 ran out of disk space and the script stopped... although it is still in the process list. We went as far as kill -9 to no avail. When we go into the directory on the linux machine where the DS1 windows share is mounted, the cursor hangs. here are the processes running... My question is, what does the ? mean in the mount.smbfs process? I am unable to umount it because it believes the mounted share is in use. Any thoughts?
Thanks!
Steve

[root@s1 /]# ps -ef | grep ds1
root 7579 1 0 Jun08 ? 00:00:00 /sbin/mount.smbfs //ds1/exps /ds1/exps -o rw workgroup HBSC username administrator
joseph 30078 1 0 Jun08 pts/5 00:00:00 ./qctool /ds1/exps/script21.spe
root 31385 30793 0 14:00 pts/15 00:00:00 grep ds1
[root@s1 /]#

 

[stevenriz]

on top of all this, this is a Windows Storage Server 2003. I tried enabling an NFS share since it has NFS services running on it. No go. As soon as I created the share we got the blue screen telling me we might be out of disk space, no driver error, just 0x0000008E ...

I thought I would try to use NFS shares instead of Windows shares for obvious reasons, but that doesn't work with windows server either!! after we free up some disk space I will give it another shot, assuming we need a certain amount of disk space for the NFS share to properly be seen from a windows platform....

 

[Kozusnik]

The only thing I'd suggest is kill -9 30078. That job may be locking it. Then umount it again.

Mark

 

[stevenriz]

yes we've tried that and it will not release. I will check each process closely to see if anything else is related... thanks for the suggestion.

 

[w33mhz]

you need to kill all the pids off then do your umount. I have ran across this myself and can be a pain on a busy directory.

 

[zeland]

Try lsof | grep -i "/ds1" to find what are the process that are holding locks on files in directory /ds1 and kill them.

--== Anything can go wrong. It's just a matter of how far wrong it will go till people think its right. ==--

 

[stevenriz]

I get this

[root@bugs ~]# lsof | grep -i "/ds1/experiments/"

and a blinking cursor. Nothing ever gets returned, even after hours....

maybe something else is wrong...

 

[thedaver]

'lsof' should end after a few seconds... run it and see if you are seeing a perpetual output... if that's the case then you have a daemon/child that is continually respawning which is probably your cause.

Can you either
1) use 'chkconfig' to disable the nfs service(s) and restart the machine, or
2) use 'killall' and name the problem service by name to kill it off?

D.E.R. Management - IT Project Management Consulting

http://www.dermanagement.com/