[etherlab-users] etherlab-users Digest, Vol 84, Issue 22

[Dave Page]

On 2014-05-29 08:33, etherlab-users-request at etherlab.org wrote:
> Thank you so much, after reading your mail, I finally understand why some
> slave goto SAFEOP+ERROR state under the circumstances. Yes I had exactly
> the same problem.
>
>

     I experience the same issue. Wiggling a marginal RJ45 will crash my 
application. In my case, the rescan traffic seemingly has priority over 
process data resulting in slave watchdog timeouts, which then cause 
further problems... This is not the best way to deal with intermittent 
failures at run time, for sure.

     I plan to make the master rescan on demand only. Once in the 
equivalent of "RUN" mode, rescans generally do not offer much value, 
outside of support for fast hot connect. The application needs to 
observe the working counter fault, return to a safe state, and then 
rescan, if required.

         Best regards - Dave Page