Re: [Testbed-admins] Capture daemon not work.

["Cheng Cui" <ccui1@tigers.lsu.edu>]

Hello,

I found the problem with the capture daemon problem in my OPS. The
$bindir in the "Setting up the Emulab Console Server" section looks has
a typo in the capture.ssh that the location of "capture" should be
"/usr/testbed/sbin/capture", not "bindir=/usr/local/sbin"

Now my capture daemon could work with the output like this:
ops# ps -aux | grep "capture"
root   30613  0.0  0.0  2472  1668  ??  Is   12:06PM   0:00.00
/usr/testbed/sbin/capture -T 15 -r -s 115200 pc1 
root   30623  0.0  0.0  2472  1668  ??  Ss   12:06PM   0:00.00
/usr/testbed/sbin/capture -T 15 -r -s 115200 pc2

However, another error message pops up when I am trying to test the
"console pc1":

ops# /usr/testbed/bin/console pc1
Traceback (most recent call last):
  File "/usr/testbed/bin/script_wrapper.py", line 2720, in <module>
    exitval  = instance.apply();
  File "/usr/testbed/bin/script_wrapper.py", line 1786, in apply
    rval,response = do_method("node", "console", params);
  File "/usr/testbed/bin/script_wrapper.py", line 214, in do_method
    ctx.load_cert(CERTIFICATE, CERTIFICATE)
  File "/usr/local/lib/python2.5/site-packages/M2Crypto/SSL/Context.py",
line 75, in load_cert
    m2.ssl_ctx_use_privkey(self.ctx, keyfile)
M2Crypto.SSL.SSLError: no start line
*** /usr/testbed/bin/console:
    /usr/testbed/bin/script_wrapper.py failed

Would anyone please give me a direction about the console setup? Thanks
a lot.

Best Regards,

Cheng Cui
. Frey Computing Services Center . Baton Rouge, LA 70803

-----Original Message-----
From: Cheng Cui [mailto:ccui1@tigers.lsu.edu] 
Sent: Friday, October 16, 2009 4:23 PM
To: 'Mike Hibler'
Cc: 'testbed-admins@flux.utah.edu'
Subject: RE: [Testbed-admins] Capture daemon not work.

Thanks to all that helped me to invoke a full path capture.

But it looks I am missing something else important which makes the
capture not run. For example, there is no daemon triggered by the
"/usr/local/etc/rc.d/capture.sh start", which I have checked with 'ops#
ps -aux | grep "*capture*"'. Nothing happens.

Then, I could see:
ops# /usr/local/etc/rc.d/capture.sh stop
No matching processes were found 

After that, I found there is no "$bindir/capture" under
"bindir=/usr/local/sbin" in the capture.ssh. 

Did I miss something?

Best Regards,

Cheng Cui
. Frey Computing Services Center . Baton Rouge, LA 70803


-----Original Message-----
From: Mike Hibler [mailto:mike@flux.utah.edu] 
Sent: Friday, October 16, 2009 1:12 PM
To: Cheng Cui
Cc: testbed-admins@flux.utah.edu
Subject: Re: [Testbed-admins] Capture daemon not work.

That example is in a really, really old FreeBSD format.  You actually
have to invoke it with the full path:

	/usr/local/etc/rc.d/capture.sh start

This will be fixed in the future.  Console setup is still a very manual
process...

On Fri, Oct 16, 2009 at 12:59:00PM -0500, Cheng Cui wrote:
> Hello,
> 
> I was testing our console server which is installed in OPS server.
After
> finished instructions for installation, I found that the
> "/usr/local/etc/rc.d/capture.sh" cannot work with the error output:
> 
> ops# ./capture.sh start
> ./capture.sh: Cannot determine the PREFIX
> 
> Did I miss something? Thanks.
> 
> Best Regards,
> 
> Cheng Cui
> . Frey Computing Services Center . Baton Rouge, LA 70803
> 
> 
> _______________________________________________
> Testbed-admins mailing list
> Testbed-admins@flux.utah.edu
> http://www.flux.utah.edu/mailman/listinfo/testbed-admins