AutoSSH problem
On Thu, May 31, 2018, 4:58 AM Germain Le Chapelain wrote: > > Hi, > > > I am experiencing troubles trying to set up a reverse SSH tunnel: > > I have installed AutoSSH as such: > > cygrunsrv.exe -I AutoSSH -p /bin/autossh -a "-M 2 -R > *:5900:localhost:5900 s...@lanvaux.fr" -y tcpip --type auto -e > AUTOSSH_NTSERVICE=yes > > > Now, it works great, except if the ssh service at lanvaux.fr would > become unavailable. > > In this case, the service stops. If you're still having problems: I've had it working for about 4 years. I've set it up on at least 4 different windows PCs. WIN7 and WIN10. I changed jobs 6 months ago so I no longer have access to any of the computers. But I did write-up my notes as a blog post: https://lizards.opensuse.org/2015/04/20/using-opensuse-as-a-reverse-tunnel-site-for-windows-7-remote-desktop/ Key section of the write-up for your issue assuming your SSHD server is Linux: === 4) Tweak the sshd_config on the server By default openSUSE enables “AllowTcpForwarding” and “TCPKeepAlive”. Verify they are either commented out or set to “yes” in /etc/ssh/sshd_config Set “GatewayPorts yes” and “ClientAliveInterval 300” in /etc/ssh/sshd_config. Also make sure they are not commented out. restart sshd to get the config values to be re-read:sudo systemctl restart sshd.service === Maybe that will help. Greg -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: AutoSSH problem
On 06/01/18 23:03, René Berber wrote: The behavior looks normal. Quoting from the autossh man page: "Startup behaviour If the ssh session fails with an exit status of 1 on the very first try, autossh 1.' will assume that there is some problem with syntax or the connection setup, and will exit rather than retrying; 2.' There is a "starting gate" time. If the first ssh process fails within the first few seconds of being started, autossh assumes that it never made it "out of the starting gate", and exits. This is to handle initial failed authentication, connection, etc. This time is 30 seconds by default, and can be adjusted (see the AUTOSSH_GATETIME environment variable below). If AUTOSSH_GATETIME is set to 0, then both behaviours are disabled: there is no "starting gate", and autossh will restart even if ssh fails on the first run with an exit status of 1. The "starting gate" time is also set to 0 when the -f flag to autossh is used. " Thank you for highlighting this point, René. You are right: it "looks" normal, from far :) However, looking closer to the log, you should see (or not, looking back at it it's not obvious) that I halted the connection the server about three minutes in: 2018/06/01 13:17:41 autossh[20680]: set alarm for 600 secs Connection to lanvaux.fr closed by remote host. 2018/06/01 13:20:52 autossh[20680]: check on child 4988 So something still doesn't line up. :/ I may try my dumpo-magic (I think it's actually called "DebugDiag", https://blogs.msdn.microsoft.com/chaun/2013/11/12/steps-to-catch-a-simple-crash-dump-of-a-crashing-process/) But as said earlier it's probably going to poop (no pun intended) about AutoSSH being a cygwin program. I will take a read on how debugging goes on Cygwin, or look at AutoSSH 1.4e (and how to get it.) Kindest regards, Thank you for your answer! -- Germain Le Chapelain Lanvaux Computer Games Limited http://lanvaux.fr -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: AutoSSH problem
On 6/1/2018 10:31 AM, Germain Le Chapelain : The behavior looks normal. Quoting from the autossh man page: "Startup behaviour If the ssh session fails with an exit status of 1 on the very first try, autossh 1.' will assume that there is some problem with syntax or the connection setup, and will exit rather than retrying; 2.' There is a "starting gate" time. If the first ssh process fails within the first few seconds of being started, autossh assumes that it never made it "out of the starting gate", and exits. This is to handle initial failed authentication, connection, etc. This time is 30 seconds by default, and can be adjusted (see the AUTOSSH_GATETIME environment variable below). If AUTOSSH_GATETIME is set to 0, then both behaviours are disabled: there is no "starting gate", and autossh will restart even if ssh fails on the first run with an exit status of 1. The "starting gate" time is also set to 0 when the -f flag to autossh is used. " -- R. Berber -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: AutoSSH problem
Hi Andrew! Thank you so much for the prompt answer, Hi. I'm glad someone is using autossh. I don't use it myself any more, and as a package maintainer it's impossible for me to tell if anyone else is. Yes, it came up first in our research to establish a reliable reverse SSH-tunnel. Is there a more common way? I figure it would be a very common problem people have, especially with things like VNC. I have also read about rstunnel ,but I don't know if it goes well on Cygwin and it was from AutoSSH so assumed AutoSSH improved on that :) SIGALRM? I wouldn't know. I was assuming some Windows thing. Actually because the retry count in the log matched 14 I assumed it had to do with that. I did glance at the AutoSSH code for a hard limit on 14 tries and indeed there are some #define of that but the default code path does seem to retry infinitely as you and I understood. As you said it seemed like it crashed anyways. Yes, it should. But in this case it looks as though the service process crashed (around 10:44). Yes! So, sorry, it doesn't look like I put that precision in my first post: The server AutoSSH was connecting to failed at that time. It only got restarted a few hours later. The first failure I was referring to right after was in the same circumstances. The server also became unavailable. (So you might even be able to reproduce the problem yourself.) I just reproduced the issue with -e AUTOSSH_DEBUG=true. The service started at 1:17 Here is the entire content of the AutoSSH.log file for the full test: 2018/06/01 13:17:41 autossh[20680]: checking for grace period, tries = 0 2018/06/01 13:17:41 autossh[20680]: starting ssh (count 1) 2018/06/01 13:17:41 autossh[4988]: child of 20680 execing /usr/bin/ssh 2018/06/01 13:17:41 autossh[20680]: ssh child pid is 4988 2018/06/01 13:17:41 autossh[20680]: check on child 4988 2018/06/01 13:17:41 autossh[20680]: set alarm for 600 secs Connection to lanvaux.fr closed by remote host. 2018/06/01 13:20:52 autossh[20680]: check on child 4988 2018/06/01 13:20:52 autossh[20680]: ssh exited with error status 255; restarting ssh 2018/06/01 13:20:52 autossh[20680]: expired child, returning 1 2018/06/01 13:20:52 autossh[20680]: checking for grace period, tries = 0 2018/06/01 13:20:52 autossh[20680]: starting ssh (count 2) 2018/06/01 13:20:52 autossh[20680]: ssh child pid is 23648 2018/06/01 13:20:52 autossh[20680]: check on child 23648 2018/06/01 13:20:52 autossh[23648]: child of 20680 execing /usr/bin/ssh 2018/06/01 13:20:52 autossh[20680]: set alarm for 410 secs ssh: connect to host lanvaux.fr port 22: Connection refused 2018/06/01 13:20:54 autossh[20680]: check on child 23648 2018/06/01 13:20:54 autossh[20680]: ssh exited with error status 255; restarting ssh 2018/06/01 13:20:54 autossh[20680]: expired child, returning 1 2018/06/01 13:20:54 autossh[20680]: checking for grace period, tries = 1 2018/06/01 13:20:54 autossh[20680]: starting ssh (count 3) 2018/06/01 13:20:54 autossh[17164]: child of 20680 execing /usr/bin/ssh 2018/06/01 13:20:54 autossh[20680]: ssh child pid is 17164 2018/06/01 13:20:54 autossh[20680]: check on child 17164 2018/06/01 13:20:54 autossh[20680]: set alarm for 409 secs ssh: connect to host lanvaux.fr port 22: Connection refused 2018/06/01 13:20:55 autossh[20680]: check on child 17164 2018/06/01 13:20:55 autossh[20680]: ssh exited with error status 255; restarting ssh 2018/06/01 13:20:55 autossh[20680]: expired child, returning 1 2018/06/01 13:20:55 autossh[20680]: checking for grace period, tries = 2 2018/06/01 13:20:55 autossh[20680]: starting ssh (count 4) 2018/06/01 13:20:55 autossh[26428]: child of 20680 execing /usr/bin/ssh 2018/06/01 13:20:55 autossh[20680]: ssh child pid is 26428 2018/06/01 13:20:55 autossh[20680]: check on child 26428 2018/06/01 13:20:55 autossh[20680]: set alarm for 408 secs ssh: connect to host lanvaux.fr port 22: Connection refused 2018/06/01 13:20:56 autossh[20680]: check on child 26428 2018/06/01 13:20:56 autossh[20680]: ssh exited with error status 255; restarting ssh 2018/06/01 13:20:56 autossh[20680]: expired child, returning 1 2018/06/01 13:20:56 autossh[20680]: checking for grace period, tries = 3 2018/06/01 13:20:56 autossh[20680]: starting ssh (count 5) 2018/06/01 13:20:56 autossh[20680]: ssh child pid is 8576 2018/06/01 13:20:56 autossh[20680]: check on child 8576 2018/06/01 13:20:56 autossh[8576]: child of 20680 execing /usr/bin/ssh 2018/06/01 13:20:56 autossh[20680]: set alarm for 407 secs ssh: connect to host lanvaux.fr port 22: Connection refused 2018/06/01 13:20:57 autossh[20680]: check on child 8576 2018/06/01 13:20:57 autossh[20680]: ssh exited with error status 255; restarting ssh 2018/06/01 13:20:57 autossh[20680]: expired child, returning 1 2018/06/01 13:20:57 autossh[20680]: checking for grace period, tries = 4 2018/06/01 13:20:57 autossh[20680]: starting ssh (count 6) 2018/06/01 13:20:57 autossh[24208]: child of 20680 execing /usr/bin/ssh 2018/06/01 13:20:57
Re: AutoSSH problem
Hi. I'm glad someone is using autossh. I don't use it myself any more, and as a package maintainer it's impossible for me to tell if anyone else is. > I have installed AutoSSH as such: > > cygrunsrv.exe -I AutoSSH -p /bin/autossh -a "-M 2 -R > *:5900:localhost:5900 s...@lanvaux.fr" -y tcpip --type auto -e > AUTOSSH_NTSERVICE=yes That all seems correct. > Now, it works great, except if the ssh service at lanvaux.fr would > become unavailable. > > In this case, the service stops. > > In the log viewer, I only get the following message in the system log: > > The description for Event ID 0 from source AutoSSH cannot be found. > Either the component that raises this event is not installed on your > local computer or the installation is corrupted. You can install or > repair the component on the local computer. > > If the event originated on another computer, the display information had > to be saved with the event. > > The following information was included with the event: > > AutoSSH: PID 8336: service `AutoSSH' failed: signal 14 raised SIGALRM? > and in autoSSH.log: > > ssh: connect to host lanvaux.fr port 22: Connection timed out > 2018/05/29 10:39:39 autossh[8424]: ssh exited with error status 255; > restarting ssh > 2018/05/29 10:40:29 autossh[8424]: starting ssh (count 12) > 2018/05/29 10:40:29 autossh[8424]: ssh child pid is 14616 > ssh: connect to host lanvaux.fr port 22: Connection timed out > 2018/05/29 10:40:50 autossh[8424]: ssh exited with error status 255; > restarting ssh > 2018/05/29 10:42:02 autossh[8424]: starting ssh (count 13) > 2018/05/29 10:42:02 autossh[8424]: ssh child pid is 20004 > ssh: connect to host lanvaux.fr port 22: Connection timed out > 2018/05/29 10:42:23 autossh[8424]: ssh exited with error status 255; > restarting ssh > 2018/05/29 10:44:01 autossh[8424]: starting ssh (count 14) > 2018/05/29 10:44:01 autossh[8424]: ssh child pid is 3056 > ssh: connect to host lanvaux.fr port 22: Connection timed out > 2018/05/29 10:44:22 autossh[8424]: ssh exited with error status 255; > restarting ssh > 2018/05/29 16:43:41 autossh[17172]: starting ssh (count 1) > > (16:43:41 is when we manually restarted the service.) > > From reading the AutoSSH manpage, I was expecting ssh to always try to > reconnect (except if it fails on the first try.) Yes, it should. But in this case it looks as though the service process crashed (around 10:44). > When I noticed the first failure, I tried changing the service settings > for AutoSSH to restart all the time, but that didn't do it. > > Am I missing something regarding AutoSSH configuration? No, it seems like a bug. To find out more, please add "-e AUTOSSH_DEBUG=true" to your service definition, and restart the service. autossh will send debugging output to autossh.log. After the service fails again, post the log output here from the last few minutes before it failed. If that doesn't help, then we may have to get more aggressive and use strace. Andrew -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
AutoSSH problem
Hi, I am experiencing troubles trying to set up a reverse SSH tunnel: I have installed AutoSSH as such: cygrunsrv.exe -I AutoSSH -p /bin/autossh -a "-M 2 -R *:5900:localhost:5900 s...@lanvaux.fr" -y tcpip --type auto -e AUTOSSH_NTSERVICE=yes Now, it works great, except if the ssh service at lanvaux.fr would become unavailable. In this case, the service stops. In the log viewer, I only get the following message in the system log: The description for Event ID 0 from source AutoSSH cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer. If the event originated on another computer, the display information had to be saved with the event. The following information was included with the event: AutoSSH: PID 8336: service `AutoSSH' failed: signal 14 raised and in autoSSH.log: ssh: connect to host lanvaux.fr port 22: Connection timed out 2018/05/29 10:39:39 autossh[8424]: ssh exited with error status 255; restarting ssh 2018/05/29 10:40:29 autossh[8424]: starting ssh (count 12) 2018/05/29 10:40:29 autossh[8424]: ssh child pid is 14616 ssh: connect to host lanvaux.fr port 22: Connection timed out 2018/05/29 10:40:50 autossh[8424]: ssh exited with error status 255; restarting ssh 2018/05/29 10:42:02 autossh[8424]: starting ssh (count 13) 2018/05/29 10:42:02 autossh[8424]: ssh child pid is 20004 ssh: connect to host lanvaux.fr port 22: Connection timed out 2018/05/29 10:42:23 autossh[8424]: ssh exited with error status 255; restarting ssh 2018/05/29 10:44:01 autossh[8424]: starting ssh (count 14) 2018/05/29 10:44:01 autossh[8424]: ssh child pid is 3056 ssh: connect to host lanvaux.fr port 22: Connection timed out 2018/05/29 10:44:22 autossh[8424]: ssh exited with error status 255; restarting ssh 2018/05/29 16:43:41 autossh[17172]: starting ssh (count 1) (16:43:41 is when we manually restarted the service.) From reading the AutoSSH manpage, I was expecting ssh to always try to reconnect (except if it fails on the first try.) When I noticed the first failure, I tried changing the service settings for AutoSSH to restart all the time, but that didn't do it. Am I missing something regarding AutoSSH configuration? It is AutoSSH 1.4f Thank you very much! Kind regards; -- Germain Le Chapelain http://lanvaux.fr -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple