dhfx

[dhfx]

I made another attempt, this time checked the server-side system logs for any clues to the apparent ssh disconnect. Here are the pertinent lines:

In server /var/log/auth.log:
Dec 29 16:04:19 D2-FRIEDMAN nxexec: pam_unix(nx:session): session opened for user friedman by (uid=116)
Dec 29 16:04:19 D2-FRIEDMAN nxexec: pam_ck_connector(nx:session): cannot determine display-device
Dec 29 16:04:19 D2-FRIEDMAN nxexec: pam_unix(nx:session): session closed for user friedman
In server /usr/NX/var/log/nxd.log:
Info: Connection from 192.168.161.4 port 47542 accepted on Mon Dec 29 16:03:59 2014.
Info: Connection from 192.168.161.4 port 47542 process 16112 started on Mon Dec 29 16:03:59 2014.
Warning: Handler process 12651 terminated by signal 1, SIGHUP.
Info: Connection from 192.168.161.7 port 56285 process 12651 closed on Mon Dec 29 16:04:16 2014.
In server /usr/NX/var/log/nxerror.log:
Info: Handler started with pid 16112 on Mon Dec 29 16:04:00 2014.
Info: Handling connection from 192.168.161.4 port 47542 on Mon Dec 29 16:04:00 2014.
Info: Connection from 192.168.161.4 port 47542 closed on Mon Dec 29 16:04:19 2014.
Info: Handler with pid 16112 terminated on Mon Dec 29 16:04:19 2014.
In server /usr/NX/var/log/nxserver.log:
2014-12-29 16:04:10 262.537 16112 NXSERVER User ‘friedman’ logged in from ‘192.168.161.4’.
2014-12-29 16:04:16 169.986 12651 NXSERVER WARNING! Process (12687) finished with signal ‘9’.
2014-12-29 16:04:16 170.095 12651 NXSERVER WARNING! NXNodeExec: session node [12687] finished abnormally with exit code [9].
Starting cleaning procedure.

The time codes should give some indication of the sequence of events. The curious thing is the message

“cannot determine display-device”

from nxexec in /var/log/auth.log. Is any of this helpful?

 

– dhfx

 

 

[dhfx]

(Britgirl: Thanks for the suggestion re disabling auto-scrolling – I will try it as soon as I get things working again.)

Since my last post I have replaced the client-side computer that suddenly died, installed my VPN (I’m using Dell NetExtender), and installed freshly-downloaded NoMachine v4.4.1_1 free (the server version is still 4.3.30 free). When I try to connect to my server using the default setup, I get a normal-looking username/password entry form, then the window goes black for several seconds, after which I see:

The connection with the server was lost.

Error is 110: Connection timed out.

I clicked on “Save logs” and in my home/Documents I find an empty NoMachine directory.

Looking directly in /usr/NX/var on the client machine I find

tail /usr/NX/var/log/nxerror.log
ssh: connect to host 127.0.0.1 port 22: Connection refused
ssh: connect to host 127.0.0.1 port 22: Connection refused

I was able to ssh into the server machine and found

tail /usr/NX/var/log/nxd.log
Info: Connection from 192.168.161.7 port 56285 accepted on Sat Dec 27 21:46:54 2014.
Info: Connection from 192.168.161.7 port 56285 process 12651 started on Sat Dec 27 21:46:54 2014.
Warning: Handler process 14079 terminated by signal 1, SIGHUP.
Info: Connection from 192.168.161.5 port 51481 process 14079 closed on Sat Dec 27 21:47:14 2014.

tail /usr/NX/var/log/nxerror.log
Info: Handler started with pid 14079 on Tue Dec 16 21:47:18 2014.
Info: Handling connection from 192.168.161.5 port 51481 on Tue Dec 16 21:47:18 2014.
Info: Connection from 192.168.161.5 port 51481 closed on Tue Dec 16 21:47:36 2014.
Info: Handler with pid 14079 terminated on Tue Dec 16 21:47:36 2014.
Info: Handler started with pid 12651 on Sat Dec 27 21:46:54 2014.
Info: Handling connection from 192.168.161.7 port 56285 on Sat Dec 27 21:46:54 2014.
Info: Connection from 192.168.161.7 port 56285 closed on Sat Dec 27 21:47:15 2014.
Info: Handler with pid 12651 terminated on Sat Dec 27 21:47:15 2014.

tail /usr/NX/var/log/nxserver.log
2014-12-27 21:47:08 976.710 12651 NXSERVER User ‘friedman’ logged in from ‘192.168.161.7’.
2014-12-27 21:47:14 774.255 14079 NXSERVER WARNING! Process (14117) finished with signal ‘9’.
2014-12-27 21:47:14 774.379 14079 NXSERVER WARNING! NXNodeExec: session node [14117] finished abnormally with exit code [9].

Starting cleaning procedure.
2014-12-27 21:47:14 781.121 12651 NXSERVER ERROR! Server process with pid finished without correct cleaning for session id

‘3A7314AA466C0CB05278479F1E7468F9’

At first glance it looks to me as if something is causing the NoMachine ssh connection to drop before the session window is filled. I would be interested to know if the logs show anything useful. If any further information would be helpful, please let me know and I will provide it. As before, thanks in advance for your help.

Again, the server is running Linux Mint 17 Cinnamon 64-bit, the client is now Linux Mint 17.1 Cinnamon 32-bit.

– dhfx

[dhfx]

Some new information: It appears that the client-screen freeze was due to a glitch in my broadband connection; a new, stronger WiMax signal came online and my modem did a reacquire, interrupting the connection for a minute or so. Interestingly, I found that the modem password had reverted to the factory default. I thought the same glitch might have affected the port forwarding, but found nothing different. In any event, t seems to me that loss of port forwarding on the client side could not have been the cause since I am connecting via a VPN which uses port 4433 (IIRC), and the VPN was still operative – I was able to ping the server machine and SSH into it after the event occurred.

I had in mind to try reinstalling 4.3.30 at the client end, but found this morning that the computer had inexplicably died during the night, so the whole matter is temporarily academic until I can replace that box. Meanwhile thanks for your attention, and I would like to take the opportunity to put in another plug for a way to disable auto-scrolling in the client window – it’s the ONE feature of NoMachine that I find highly annoying.

 

– dhfx

[dhfx]

Tor: Thanks for the quick reply. A checkbox would be ideal, although I would be happy to have it accessible in the config file, since it is more in the line of a user preference (of course, this implies that for multiuser systems, there is a separate config file for each user).

A refinement would be to be able to specify the depth of the sensitive region, i.e., the extent in pixels from the window edge. A depth of 1 would still allow the viewport to be moved by “pushing” the edge of it with the mouse pointer, and a depth of 0 or -1 would in effect disable the auto-scrolling completely.

BTW, is there a list of existing options that are controlled from the config file?

– dhfx

 

[Author]

Posts