So I'm working with RANCID to monitor some switches and I'm using clogin now to test some modifications.
I'm working in two different computers but both of them from the office. Althought they are on different subnets, both can stablish telnet and ssh connections to the switch I'm working with right now (Alcatel Omni 6248). I've made the default installation of RANCID on both computers and added everything needed.
Don't bother thinking of a missconfiguration yet
So I managed to get the config for some of the switches but two of them failed so I tried to modified clogin so it gets every router configuration. After some changes, executing
./clogin -c "show running-config" alcatel-sw
would lead into the script loging in, executing
terminal datadump
and then waiting on the prompt untill it gets a TIMEOUT. While trying to fix it, I just decided to get the default clogin and start from zero again so I recovered the backup copy and tried to log into a CISCO switch (which perfectly works with the default clogin script). I executed
./clogin -c "show running-config" cisco-sw
and I got a TIMEOUT as if I were using the old script. I checked it and nope, I was using the default one. I even transfered the one I have on the other computer (which I tested before sending) and tried again getting same error.
Any idea on how can this be possible?
I'm on CentOS 7.2 on same machines, both on an almost fresh installation and clogin was working untill I modified it and then replaced it with the backup copy.
This is the second time I waste two days of my live looking for error while it was SElinux the whole time.
Basically, when doing
cp clogin clogin.bk, it was missing the context so it was not able to fully work.It works fine again after doing:
chcon unconfined_u:object_r:admin_home_t:s0 clogin