Terminating a subprocess question

[Ernesto]

I'm opening a telnet session with the subprocess call below. I then
wait ten seconds and attempt to terminate the telnet window I created.
Unfortuantely, the telnet window remains after the 'TerminateProcess"
call below. This software works great for opening an executable
directly (i.e. Handle = subprocess.Popen("myProgram.exe), but here I'm
using the Windows 'start' command inside, which I think has an effect.
Is there another way I can force termination of the telnet session I
create ?

# CODE STARTS HERE

TSS_Log_Path = "C:\\Log_Outputs\\TSS_Log_Test.txt"
TSS_Handle = subprocess.Popen("start telnet.exe -f " + TSS_Log_Path + "
localhost 6000",shell=True)
time.sleep(10)
ctypes.windll.kernel32.TerminateProcess(int(TSS_Ha ndle._handle), -1) #
Terminate the TSS_Log

# END CODE

[Diez B. Roggisch]

Ernesto wrote:

I'm opening a telnet session with the subprocess call below. I then
wait ten seconds and attempt to terminate the telnet window I created.
Unfortuantely, the telnet window remains after the 'TerminateProcess"
call below. This software works great for opening an executable
directly (i.e. Handle = subprocess.Popen("myProgram.exe), but here I'm
using the Windows 'start' command inside, which I think has an effect.
Is there another way I can force termination of the telnet session I
create ?

Maybe using the module telnetlib instead?

Diez

[Ernesto]

Diez B. Roggisch wrote:

Ernesto wrote:

I'm opening a telnet session with the subprocess call below. I then
wait ten seconds and attempt to terminate the telnet window I created.
Unfortuantely, the telnet window remains after the 'TerminateProcess"
call below. This software works great for opening an executable
directly (i.e. Handle = subprocess.Popen("myProgram.exe), but here I'm
using the Windows 'start' command inside, which I think has an effect.
Is there another way I can force termination of the telnet session I
create ?

Maybe using the module telnetlib instead?

Diez

oh ok great ! But how would I create a log file. Couldn't seem to
find that here:

http://python.active-venture.com/lib...t-objects.html

http://pydoc.org/1.6/telnetlib.html

Plus, everytime, I tried to use the read_all object, my program crashed
pretty hard...

tn = telnetlib.Telnet("localhost",6000)
print tn.read_all()
# CRASH

[Fredrik Lundh]

Ernesto wrote:

Plus, everytime, I tried to use the read_all object, my program crashed
pretty hard...

tn = telnetlib.Telnet("localhost",6000)
print tn.read_all()
# CRASH

that's an unusual error message. are you sure you didn't get a
traceback? if so, what did it say?

</F>

[Ernesto]

Fredrik Lundh wrote:

Ernesto wrote:

Plus, everytime, I tried to use the read_all object, my program crashed
pretty hard...

tn = telnetlib.Telnet("localhost",6000)
print tn.read_all()
# CRASH

that's an unusual error message. are you sure you didn't get a
traceback? if so, what did it say?

</F>

I was running it directly in a python shell. After the tn.read_all()
call, the python shell window freezes up, and I have to do a hard
termination of the shell. There is no traceback message, just a freeze.

[Fredrik Lundh]

Ernesto wrote:

tn = telnetlib.Telnet("localhost",6000)
print tn.read_all()
# CRASH

that's an unusual error message. are you sure you didn't get a
traceback? if so, what did it say?

I was running it directly in a python shell. After the tn.read_all()
call, the python shell window freezes up, and I have to do a hard
termination of the shell. There is no traceback message, just a freeze.

hmm. that hardly qualifies as a "CRASH"; rather, it's pretty much
what you'd expect from a read_all call:

help(tn.read_all) Help on method read_all in module telnetlib:

read_all(self) method of telnetlib.Telnet instance
Read all data until EOF; block until connection closed.

(note the "block until connection closed" part)

what happens if you use a non-blocking method instead ? e.g.
help(tn.read_very_eager)

Help on method read_very_eager in module telnetlib:

read_very_eager(self) method of telnetlib.Telnet instance
Read everything that's possible without blocking in I/O (eager).

Raise EOFError if connection closed and no cooked data
available. Return '' if no cooked data available otherwise.
Don't block unless in the midst of an IAC sequence.

</F>

[Ernesto]

Ernesto wrote:

I'm opening a telnet session with the subprocess call below. I then
wait ten seconds and attempt to terminate the telnet window I created.
Unfortuantely, the telnet window remains after the 'TerminateProcess"
call below. This software works great for opening an executable
directly (i.e. Handle = subprocess.Popen("myProgram.exe), but here I'm
using the Windows 'start' command inside, which I think has an effect.
Is there another way I can force termination of the telnet session I
create ?

# CODE STARTS HERE

TSS_Log_Path = "C:\\Log_Outputs\\TSS_Log_Test.txt"
TSS_Handle = subprocess.Popen("start telnet.exe -f " + TSS_Log_Path + "
localhost 6000",shell=True)
time.sleep(10)
ctypes.windll.kernel32.TerminateProcess(int(TSS_Ha ndle._handle), -1) #
Terminate the TSS_Log

# END CODE

Actually, the original answer I was looking for to for a "hard close"
of telnet in Windows is:

TSS_Handle = subprocess.Popen("TASKKILL /F /IM telnet.exe", shell=True)

It is almost definitely true though that it is safer to use telnetlib.
I just don't have the time to adjust my entire huge application at the
moment. Thanks all.