Now it seems working, graphs are printed, yeah!
anyway this is not what I expected, since even if I'm really able now to
filter by hostname, I receive a graph which contains all the line values
for each host and below the hostnames (that I don't want to be printed).
What I need is the average total for these filtered nodes, like it's
showed for the whole grid in the main page (for example).
I need only the total average graph :)
thanks for all
cheers,
Michele
Il giorno gio, 24/11/2011 alle 11.59 +0000, Gerhard Lazu ha scritto:
> That looks good to me.
>
>
> My URL for the routers CPU aggregate looks like this:
>
>
> graph_all_periods.php?hreg[]=router&mreg[]=cpu_%28system%7Cuser%
> 29>ype=line&vl=Percent&title=Routers%20CPU&aggregate=1&z=large
>
> Just to double-check, you are using the actual hostname for the
> host_regex, right? I'm overriding my routers' hostname to router1,
> router2, etc. via gmond.conf, hence the "router" regex correctly
> identifies them all. They are all EC2 instances that come and go based
> on auto-scaling rules. This way, I'm ensuring that the graphs have
> continuity for router1, router2 etc. even if the instances might not
> be the same ones.
>
>
> Gerhard
>
>
> ______________________________________________________________________
> Twitter Github Blog
>
>
> On Thu, Nov 24, 2011 at 11:33 AM, mcarpene <m.car...@cineca.it> wrote:
>
>
> Thanks a lot Gerhard,
> I've tested your example, but I cannot visualize correctly the
> graph in
> the view:
>
> I see that the image generated cannot be displayed (maybe not
> created
> correctly)
>
> looking at the url I see:
>
> /graph_all_periods.php?hreg[]=node_name&mreg[]=cpu_%28system|
> user%
> 29>ype=line&vl=Percent&title=CLUSTER%
> 20CPU&aggregate=1&z=large
>
> the file I modified is
>
> {
> "view_name": "xxx_cluster",
> "items": [
> {
> "vertical_label": "Percent",
> "title": "CLUSTER CPU",
> "metric_regex": [
> {
> "regex": "cpu_(system|user)"
> }
> ],
> "graph_type": "line",
> "aggregate_graph": "true",
> "host_regex": [
> {
> "regex": "node_name"
> }
> ]
> },
> {
> "vertical_label": "Load",
> "title": "CLUSTER Load",
> "metric_regex": [
> {
> "regex": "load_one"
> }
> ],
> "graph_type": "line",
> "aggregate_graph": "true",
> "host_regex": [
> {
> "regex": "node_name"
> }
> ]
> }
> ],
> "view_type": "standard"
> }
>
>
> I think this
>
> hreg[]=node_name
>
> is something not parsed correctly.
>
> thank you for any information.
>
> cheers,
> Michele
>
>
>
> Il giorno mer, 23/11/2011 alle 22.46 +0000, Gerhard Lazu ha
> scritto:
> > Aggregate view, taken from a live ganglia v3.2.0 & gweb
> v2.1.8.
> >
> >
>
> >
> ______________________________________________________________________
> > Twitter Github Blog
> >
> >
> > On Wed, Nov 23, 2011 at 10:44 AM, mcarpene
> <m.car...@cineca.it> wrote:
> > Hi all,
> > I'm using last ganglia web portal version: 2.1.8.
> > I need to create a simple custom view, containing
> average data
> > for load
> > and cpu about a single cluster.
> > The problem is that I've to visualize the summary
> data about
> > only a
> > certain number of nodes (not all) and I don't know
> if that's
> > possible:
> > I'm trying creating a special view like this:
> >
> > {
> > "view_name":"cluster_x",
> > "items":[
> > {
> > "metric_regex":[{"regex":"load_one"}],
> > "host_regex":[{"regex":"regular_expression"}],
> > "graph":"load_report"
> > },
> > {
> > "metric_regex":[{"regex":"cpu_report"}],
> > "host_regex":[{"regex":"regular_expression"}],
> > "graph":"cpu_report"
> > }
> > ],
> > "view_type":"standard"
> > }
> >
> > two graphs appear, but they don't show correct data,
> the first
> > one is
> > empty at all.
> >
> > do you have any example?
> >
> > Regards,
> > Michele
> >
> >
> >
>
> ------------------------------------------------------------------------------
> > All the data continuously generated in your IT
> infrastructure
> > contains a definitive record of customers,
> application
> > performance,
> > security threats, fraudulent activity, and more.
> Splunk takes
> > this
> > data and makes sense of it. IT sense. And common
> sense.
> > http://p.sf.net/sfu/splunk-novd2d
> > _______________________________________________
> > Ganglia-general mailing list
> > Ganglia-general@lists.sourceforge.net
> >
> https://lists.sourceforge.net/lists/listinfo/ganglia-general
> >
>
>
>
>
>
------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
Ganglia-general mailing list
Ganglia-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ganglia-general
