[rancid] Device not contacted

[Alan McKinnon]

Hi Ryan,

Run this manually:

nlogin -t 90 -c"get system;get conf" <firewall address> | less

It's the same command that nrancid will run. You can pipe the output to
less, or redirect it to a file or whatever, the idea here is to be able
to examine the output at leisure.

"missed cmds" always means little more than "something went wrong", it's
your signal to sharpen those debugging tools :-)

I'm not familiar with NetScreens, but there's only two commands run by
nrancid so it gets easier. Look at the nlogin output above carefully
with fresh eyeballs. At the point where get system ends and get conf
starts, is there anything odd there? Is the prompt correct?

The code that finds the start of each command in the output and runs the
correct sub is this (edited for clarity):


TOP: while(<INPUT>) {
    while (/>\s*($cmds_regexp)\s*$/) {
        $cmd = $1;
        if (!defined($prompt)) {
            $prompt = ($_ =~ /^([^>]+->)/)[0];
            $prompt =~ s/([][}{)(\\])/\\$1/g;
            print STDERR ("PROMPT MATCH: $prompt\n") if ($debug);
        }
        $rval = &{$commands{$cmd}};
        delete($commands{$cmd});
        if ($rval == -1) {
            last TOP;
        }
    }
}

and the sub does this (also edited for clarity):

sub GetSystem {

    while (<INPUT>) {
        last if(/$prompt/);
    }
}

Basically, if the prompt doesn't match what the script expects it to be,
GetSystem runs till the very end, GetConf never runs at all, and "get
conf" is tagged as not being detected - the error you see.

The prompt must match the regexes in the "if (!defined($prompt))" for
all of this magic to work at all.

If nothing seems untoward at this point, run:

nrancid -d <firewall address> and look for the text "PROMPT MATCH",
verify that it matches reality.

There can be other reasons for this error, but lets go for the common
low hanging fruit first.




On 17/12/2013 17:39, Ryan Milton wrote:
> Hi all (Alan),
> 
>  
> 
> Rancid has generally been running smooth as can be with my small
> deployment. I have just added two more Netscreen firewalls, to bring to
> a total of 6 Netscreen204s.
> 
>  
> 
> Only issue is that one of the new devices is being a pain with that
> dreaded “The following routers have not been successfully contacted for
> more than 4 hours.” Error.
> 
>  
> 
> All  devices have same login in the .clonginrc
> 
>  
> 
> All devices can be logged into manually using the nlogin script.
> 
>  
> 
> Log files only show:
> 
> Getting missed routers: round 1.
> 
> 195.3.XXX.XXX: missed cmd(s): get conf
> 
> 195.3.XXX.XXX: End of run not found
> 
> #
> 
>  
> 
> The router.db file is fine, and the FW has the correct login info. Any
> idea what might the issue be?
> 
>  
> 
> Regards,
> 
> Ryan Milton
> 
> MVS Network Manager
> 
> o: 201-447-1505 x124
> 
> c: 862-249-5230
> 
> www.mvsusa.com <http://www.mvsusa.com/>
> 
> MVS final logo GOOD very small
> 
>  
> 
>  
> 
> 
> 
> _______________________________________________
> Rancid-discuss mailing list
> Rancid-discuss at shrubbery.net
> http://www.shrubbery.net/mailman/listinfo/rancid-discuss
> 


-- 
Alan McKinnon
alan.mckinnon at gmail.com