TransportError: Not connected to NETCONF server about ansible-junos-stdlib HOT 4 CLOSED

Did more digging on this. Turns out that this is related to the error above, and the error is dependent upon the configuration contents. This device has two sections of config: system and snmp. Sending both sections together fails. Sending one section at a time (system but no snmp, or snmp but no system) works fine. So it is not that the config contains anything offensive, but perhaps that there's a length (buffer?) issue. Closest thing I found reported was:

http://forums.juniper.net/t5/SRX-Services-Gateway/strange-NETCONF-error-any-ideas-how-to-debug-the-root-cause/td-p/201349

But no solution was offered.

Again, any ideas on how to approach this would be very appreciated.

from ansible-junos-stdlib.

Another point is that we have many more EX-4200 units, with the exact same configuration being pushed via netconf, but only these two are running 12.3R4.6.

from ansible-junos-stdlib.

I noticed that the errors started after we added a user account. It turns out that shortening the user's full name makes the error go away. This feels like a Junos bug, but I wasn't able to find anything in the Changelog for 12.3

Anyways, I'm closing this one for now. Hopefully this monologue will be helpful to someone else :-)

from ansible-junos-stdlib.

This behavior and the error message is similar to a case we have identified some times in the past but not in a deterministic way, therefore we ignored it. Now we managed to reproduce it. See ncclient/ncclient#189

from ansible-junos-stdlib.