Hello Martin, It is a good news to see that someone take care to retrieve this 3.0.x feature in 3.1.x. Agree with you, need to have also the SPOOF_HOST in the call_back !
BTW, let me give you my feed back on usage of the spoof option in 3.0.x. >From my point of view, we need to share two kids of usage. 1) The users expect to spoof only on Ganglia know host and up. ---> In such a case the front-end doesn't need any patch 2) The users expect to spoof on host unknow of Ganglia, mean's host where gmond is not running. ---> In such a case the front-end need at least some patch to handle properly such kind of strange host... >From my understanding, in 3.0.x, Ganglia doesn't take care of this two different cases. Then the first question is : Does it was the goal of the "spoof" feature to handle the two cases ? because in 3.0.x we can't say that the case 2) is fully supported by Ganglia ( front-end need patch ). In my company, we only deal with 1), but it sound that some other ones use 2) .... In 3.0.x, when we spoof on a down host or unknown host ( at gmond network point of view ), then this host is displayed in the front-end as down. To avoid that, in 3.0.x we need to check before spoofing a host that this one is not down, using gstat it is not a big stuff.... Nevertheless, it will be nice if this test would be embedded in Ganglia, with a drop of all the metrics spoofed on a down host or unknown host, like that no any additional test will be needed before sending metrics. But for that Ganglia need to know at least if we are in 1) or 2), because in case 2) the users don't want that Ganglia drop the metric. Maybe a configuration flag for case 1) or 2) would help Ganglia to know what behaviour to apply ? Regards. Christian. ----- Original Message ----- From: "Christian Gouret" <[EMAIL PROTECTED]> To: "Martin Hicks" <[EMAIL PROTECTED]>; <[email protected]> Sent: Sunday, October 05, 2008 10:22 AM Subject: Re: [Ganglia-general] metric name and spoofed metrics > Hello Martin, > It is a good news to see that someone take care to retrieve this 3.0.x > feature in 3.1.x. > Agree with you, need to have also the SPOOF_HOST in the call_back ! > > BTW, let me give you my feed back on usage of the spoof option in 3.0.x. > From my point of view, we need to share two kids of usage. > 1) The users expect to spoof only on Ganglia know host and up. > ---> In such a case the front-end doesn't need any patch > 2) The users expect to spoof on host unknow of Ganglia, mean's host where > gmond is not running. > ---> In such a case the front-end need at least some patch to handle > properly such kind of strange host... > > From my understanding, in 3.0.x, Ganglia doesn't take care of this two > different cases. > > Then the first question is : Does it was the goal of the "spoof" feature > to handle the two cases ? > because in 3.0.x we can't say that the case 2) is fully supported by > Ganglia ( front-end need patch ). > > In my company, we only deal with 1), but it sound that some other ones use > 2) .... > > In 3.0.x, when we spoof on a down host or unknown host ( at gmond network > point of view ), then this host is displayed in the front-end as down. > To avoid that, in 3.0.x we need to check before spoofing a host that this > one is not down, using gstat it is not a big stuff.... > Nevertheless, it will be nice if this test would be embedded in Ganglia, > with a drop of all the metrics spoofed on a down host or unknown host, > like that no any additional test will be needed before sending metrics. > But for that Ganglia need to know at least if we are in 1) or 2), because > in case 2) the users don't want that Ganglia drop the metric. > Maybe a configuration flag for case 1) or 2) would help Ganglia to know > what behaviour to apply ? > Regards. > Christian. > > > ----- Original Message ----- > From: "Martin Hicks" <[EMAIL PROTECTED]> > To: <[email protected]> > Sent: Thursday, October 02, 2008 3:05 PM > Subject: [Ganglia-general] metric name and spoofed metrics > > >> >> >> (sorry if this is a duplicate. I sent it yesterday but I haven't seen >> it come back yet, nor has it shown up in the mailing list archives on >> sourceforge) >> >> Hi, >> >> I backported the spoofing patches to my 3.1.1 build (I also check this >> against trunk to make sure I hadn't missed something) in order to play >> with python DSO and spoofing. >> >> I ran into a question. If you have the same metric for a bunch of >> different hosts, just with different SPOOF_HOST, then how do you tell >> the difference in the call_back? You just get the 'name', which appears >> to always be the same. >> >> Here's my test example: >> >> import pprint >> import os >> >> def metric_init(params): >> global descriptors >> >> metrics = [] >> metric = {'name': 'myMetric', >> 'call_back': metric_handler, >> 'time_max': 90, >> 'value_type': 'uint', >> 'units': 'units', >> 'format':'%u', >> 'description': 'Spoofed Metric', >> 'groups':'testmort'} >> >> spoof_1 = metric >> spoof_1['SPOOF_NAME'] = 'myMetric-r1i0n0' >> spoof_1['SPOOF_HOST'] = "192.168.159.10:r1i0n0" >> >> spoof_2 = metric >> spoof_2['SPOOF_NAME'] = "myMetric-r1i0n8" >> spoof_2['SPOOF_HOST'] = "192.168.159.18:r1i0n8" >> >> metrics = [spoof_1, spoof_2] >> pprint.pprint(metrics) >> descriptors = metrics >> return descriptors >> >> >> def metric_cleanup(): >> pass >> >> >> def metric_handler(name): >> print "name = '%s'" % name >> return 1 >> >> >> >> >> Am I expected to deal with each SPOOF_HOST when a call_back occurs for a >> particular metric name? >> >> I kind of expected SPOOF_NAME to be in the 'name' argument of the >> call_back. >> >> thanks >> mh >> >> ------------------------------------------------------------------------- >> This SF.Net email is sponsored by the Moblin Your Move Developer's >> challenge >> Build the coolest Linux based applications with Moblin SDK & win great >> prizes >> Grand prize is a trip for two to an Open Source event anywhere in the >> world >> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> _______________________________________________ >> Ganglia-general mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/ganglia-general >> > ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Ganglia-general mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ganglia-general
