[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[LDM #RCA-822786]: Failover doesn't occur when primary feed goes down

Gilbert,

> > I didn't see anything in the log file to indicate that the LDM on
> >Weather2 was ever requesting data-products of feedtype "NEXRAD3|WMO"
> >(which is the same as "IDS|DDPLUS|HRS|NNEXRAD") from
> >"idd.unidata.ucar.edu". Can you find a "Starting Up" message to that
> >effect in the earlier log files? The command
> >
> >>    fgrep 'Starting Up' `ls -tr ldmd.log*` | fgrep WMO | fgrep
> >idd.unidata.ucar.edu | tail
> >
> > should do it.
> 
> I went through it this morning. You are right...and I have fixed that now.
> Also...
> 
> > I did notice partially overlapping requests regarding the FNEXRAD feedtype. 
> > You should vet the LDM configuration-file regarding that feedtype. Feeds 
> > should be disjoint or identical: partial overlaps in the feedtype/pattern 
> > specification can result in missed data-products.
> 
> Yep. I had put FNEXRAD in a request of a group of feeds, and then I
> put it later on down the file as a single request of that feed. I
> commented the latter out and restarted the LDM.
> 
> > If you'll send me Weather2's LDM configuration-file I'll be happy to
> > take a look at it.
> 
> I think I am good now...I'd rather you spend the time to figure out why
> the backup feed dies when it's supposed to fail over from time to time.
> And if there's anything I can do to help with that, please let me know!

Looking at the configuration-file supports the fail-over investigation. If 
there was no downstream LDM making the identical request for NEXRAD3|WMO 
data-products from idd.unidata.ucar.edu as was made to Weather3, then there was 
no process that could take-over when Weather 3 went offline.

Regards,
Steve Emmerson

Ticket Details
===================
Ticket ID: RCA-822786
Department: Support LDM
Priority: Normal
Status: Closed