Oleg, I don't know specifically about a limit on the nodes, but I'm curious as to why you have this many nodes listed for a single data source? Can these nodes not see each other via multicast? Assuming they are part of the same multicast cluster, then any single node should know the status of all nodes in the cluster. All you really need is one node listed in your gmetad.conf (include one or two extra for failover), and everything should work for you...gmetad will get the status of the entire cluster from a single node. No need to list them all.
Steve Gilbert Unix Systems Administrator [EMAIL PROTECTED] -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Oleg Lodygensky Sent: Tuesday, November 23, 2004 8:14 AM To: [email protected] Subject: [Ganglia-general] nodes per cluster hello, sorry if this has already been posted but I can't find any clue on the mail list archive. I'm under linux debian (kernel 2.6.9) Using ganglia 2.5.6 gcc --version = gcc (GCC) 3.3.5 (Debian 1:3.3.5-2) Is there a limit on the number of nodes per cluster? In my gmetad.conf my data source is defined as : data_source "mycluster" dummy00.nowhere.com:8694 dummy01.nowhere.com:8694 dummy02.nowhere.com:8694 dummy03.nowhere.com:8694 dummy04.nowhere.com:8694 dummy05.nowhere.com:8694 dummy06.nowhere.com:8694 dummy07.nowhere.com:8694 dummy08.nowhere.com:8694 dummy09.nowhere.com:8694 dummy10.nowhere.com:8694 dummy11.nowhere.com:8694 dummy12.nowhere.com:8694 dummy13.nowhere.com:8694 dummy14.nowhere.com:8694 dummy15.nowhere.com:8694 dummy16.nowhere.com:8694 dummy17.nowhere.com:8694 dummy18.nowhere.com:8694 dummy19.nowhere.com:8694 dummy20.nowhere.com:8694 dummy21.nowhere.com:8694 dummy22.nowhere.com:8694 dummy23.nowhere.com:8694 dummy24.nowhere.com:8694 dummy25.nowhere.com:8694 dummy26.nowhere.com:8694 dummy27.nowhere.com:8694 dummy28.nowhere.com:8694 dummy29.nowhere.com:8694 dummy30.nowhere.com:8694 dummy31.nowhere.com:8694 dummy32.nowhere.com:8694 dummy33.nowhere.com:8694 dummy34.nowhere.com:8694 dummy35.nowhere.com:8694 dummy36.nowhere.com:8694 dummy37.nowhere.com:8694 dummy38.nowhere.com:8694 dummy39.nowhere.com:8694 dummy40.nowhere.com:8694 dummy41.nowhere.com:8694 And gmetad has a Segmentation fault; if I remove the 42nd and last host (dummy41), everything is just ok; even if here, of course, these hosts do certainly not exist... :) I tried to define my data source like this: data_source "mycluster" dummy00.nowhere.com:8694 dummy01.nowhere.com:8694 dummy02.nowhere.com:8694 dummy03.nowhere.com:8694 dummy04.nowhere.com:8694 dummy05.nowhere.com:8694 dummy06.nowhere.com:8694 dummy07.nowhere.com:8694 dummy08.nowhere.com:8694 dummy09.nowhere.com:8694 \ dummy10.nowhere.com:8694 dummy11.nowhere.com:8694 dummy12.nowhere.com:8694 dummy13.nowhere.com:8694 dummy14.nowhere.com:8694 dummy15.nowhere.com:8694 dummy16.nowhere.com:8694 dummy17.nowhere.com:8694 dummy18.nowhere.com:8694 dummy19.nowhere.com:8694 \ dummy20.nowhere.com:8694 dummy21.nowhere.com:8694 dummy22.nowhere.com:8694 dummy23.nowhere.com:8694 dummy24.nowhere.com:8694 dummy25.nowhere.com:8694 dummy26.nowhere.com:8694 dummy27.nowhere.com:8694 dummy28.nowhere.com:8694 dummy29.nowhere.com:8694\ dummy30.nowhere.com:8694 dummy31.nowhere.com:8694 dummy32.nowhere.com:8694 dummy33.nowhere.com:8694 dummy34.nowhere.com:8694 dummy35.nowhere.com:8694 dummy36.nowhere.com:8694 dummy37.nowhere.com:8694 dummy38.nowhere.com:8694 dummy39.nowhere.com:8694\ dummy40.nowhere.com:8694 but this changed nothing (as expected since CFG_BUFSIZE = 4096 [dotconf.h] ) any clue ? thanx, Oleg Lodygensky ------------------------------------------------------- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/ _______________________________________________ Ganglia-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ganglia-general
