it requires patches from earlier bugs #66, #68 applied or the patch will not apply cleanly.
also as it is kind of rewrite of whole plugin, i suppose new version 1.2 be called out :)
also currently query timeout is hardcoded as 15sec, maybe make it also configurable (it is different than global script timeout!).
however due the way script works, if it queries 20 servers at at time (worker count) and 19 of them timeouts, then the total timeout for whole script is still 15sec, but if you have 40 servers to query, but worker size 20, and 39 of servers timeout, the whole script timeout is 30sec....
so maybe use the existing --timeout param for that 15sec and suggest to use bigger worker count.... or redo the script and always use worker count = number of servers to query...
Original reporter: Elan Ruusamäe glen@delfi.ee
i've coded a patch just now that does parallel dns lookups instead of forking. the result is achieved even faster (likely due less cpu usage).
it's probably experimental as it is so fresh and hot, but i'd be happy to get it applied or feedback :)
http://cvs.pld-linux.org/cgi-bin/cvsweb.cgi/packages/nagios-plugin-check_rbl/mdns.patch
it requires patches from earlier bugs #66, #68 applied or the patch will not apply cleanly.
also as it is kind of rewrite of whole plugin, i suppose new version 1.2 be called out :)
also currently query timeout is hardcoded as 15sec, maybe make it also configurable (it is different than global script timeout!).
however due the way script works, if it queries 20 servers at at time (worker count) and 19 of them timeouts, then the total timeout for whole script is still 15sec, but if you have 40 servers to query, but worker size 20, and 39 of servers timeout, the whole script timeout is 30sec....
so maybe use the existing --timeout param for that 15sec and suggest to use bigger worker count.... or redo the script and always use worker count = number of servers to query...