[LON-CAPA-admin] Cron Job

H. K. Ng ng at martech.fsu.edu
Mon Jan 24 10:24:35 EST 2005


Hi Guy,


> > >
> > >version 1.3.2 will be much better about this.
> > >
> > >There was a 'sleep 1;' in loncnew that was causing havoc with the
> > >'pong' command.
> > >
> > >Which only really gets used by loncron. Which will also stop using it
> > >in 1.3.2
> >
> >
> > Does this fix the long delay in the republication of problems?
>
>I sure hope it does, Gerd was complaining about similar effects right
>when Mark posted hi complaint.
>
>If you have a chance to upgrade your cluster to the lastest version of
>loncnew and version 1.587.2.1 of lonnet.pm I'd love to hear if things
>improve.

I updated the fsu and fsuK12 clusters and the republication is now 
significantly faster - like a few seconds (for one test). However,

ps auxwww | grep lonc

still shows lots of "DEAD" connections to the outside and some does not 
show connection between servers in the fsu and fsuK12 clusters. It does not 
affect logins though??

-hk




More information about the LON-CAPA-admin mailing list