Weird behaviour from argus

[Carter Bullard]

Hey Russell,
Yes, Peter and I are working on this very problem, and in the process
have generated quite a bit of dust, but we're getting it cleaned up.
If you can upload the file, I am working on putting automatic recovery
in the file data processing so we can recover from this error (basically
we're putting out some zero's when we're not really suppose to, and
good records start again at the end of the string of zeros, at least  
that
is the behavior I've seen so far).

Sorry, we should have an update hopefully very soon!!!

Carter


On Aug 15, 2007, at 5:42 PM, Russell Fulton wrote:

> I have some weird problems, this (or something very similar) is
> happening on two sensors.  When I run ra on a file (any file) I get  
> back
> just two records and then get the error:
>
>  ra[20871]: 09:18:05.640627 ArgusReadStreamSocket (0xb740d954) record
> length is zero
>
> either on the console or (as in this case) in syslog.
>
>
> In both cases I am running argus-3.0.0 as the server and the RC46 and
> RC39 clients.
>
> Luckily (or was it good planning ;) I have other sensors on the  
> path of
> the traffic I was tracking and this on is working fine.
>
> I admit not paying much attention to the list traffic recently so my
> apologies if this has come up and I missed it.
>
> Carter: I can upload one of these files to the ftp server if you  
> want to
> do diagnostics.
>
> OH yes! the platform is FC6.
>
> Russell.
>