nice idea but I really need nearly realtime handling of perfdata and
oc(s|h)p commands. 20seconds are much too much.=20
As for me I won't care about a little bit of data loss if I get a
warning message inside the logs (like running out of external command
buffer slots) or maybe via check_nagios
. =20I hold that data loss because of a full pipe shouldn't be a normal case.
In case it happens something is wrong and has to be fixed. =20
Gerd=20
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]