| Title: | DIGITAL UNIX (FORMERLY KNOWN AS DEC OSF/1) |
| Notice: | Welcome to the Digital UNIX Conference |
| Moderator: | SMURF::DENHAM |
| Created: | Thu Mar 16 1995 |
| Last Modified: | Fri Jun 06 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 10068 |
| Total number of notes: | 35879 |
Hi, I got the problem that a process does not react on signals as KILL TRAP, .... - the process never returns from kernel mode to receive and react on the signal. Q: Is there a way to kill a process by KDBX or DBX -K ??? Thanks for the help ---Maik [Posted by WWW Notes gateway]
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 9039.1 | SMURF::DENHAM | Digital UNIX Kernel | Wed Mar 05 1997 10:55 | 9 | |
The simple answer is no. If SIGKILL doesn't work, nothing will.
Now, for the facts. This is a plain old process? If so, we need
to know why its stuck in the kernel. Get us a kernel stack
trace to look (dbx -k /vmunix; set $pid=XXX; tstack).
If it's some kind of driver or subsystem thing, then the rules
are different. I need to know how the procress/thread was created
and so on.
| |||||
| 9039.2 | at least that's how I understand it... | SMURF::PETERT | rigidly defined areas of doubt and uncertainty | Wed Mar 05 1997 13:05 | 11 |
If you do a 'ps x' does the state show up as 'U'? This would mean
it is uninterruptable, and is waiting on some resource that is
either not available, or is tied up by some other process.
In this case, there is not much you can do about it, much as Jeff
replied in the previous note. If it is in this state, it shouldn't
be using any cycles, just stuck at this point. Unless the resource
suddenly becomes available, the process will hang in this state
until the system reboots.
PeterT
| |||||