Thanks Chris. Bsically i am using vcmts as my VNF and I'm looking to switch from collectd to the node_exporter container for a telemetry related project. Everything works great, except for the fact that we need to collect more perf events than are currently enabled in the default container. Is there a way to pass a perf event list into node exporter somehow? I noticed that the events are defined in the perf_linux.go <https://github.com/prometheus/node_exporter/blob/master/collector/perf_linux.go> file. This means I would have to hack the events I need into the file and rebuild bu nothing related to LLC-stores and also pushing vcmts traffic metrics into node-exporter . Any thoughts?
Thanks, Divya On Tuesday, June 16, 2020 at 12:58:43 PM UTC+5:30, Christian Hoffmann wrote: > > Hi, > > On 6/16/20 7:42 AM, Divya Rajasekaran wrote: > > Is there any way to push 3rd party VNF metric into the node exporter > > Are you referring to Virtual Network Functions as described here? > https://tools.ietf.org/id/draft-rosa-bmwg-vnfbench-01.html > > I don't think node_exporter has support for this. > > Could this be what you are looking for? > https://osm.etsi.org/wikipub/index.php/OSM_Performance_Management > > If this doesn't help, you might have to look for another solution. If > this is only about some additional machine-specific metrics, it might be > a good fit for a pull request for node_exporter (check with the > maintainers first). A simpler approach would be using the > node_exporter's textfile collector for that and exporting the required > metrics to a "vnf.prom" file or something. > > If this is something bigger / non-machine specific, it might make sense > to use separate exporter. > > > and also I am not able to see LLC-stores metrics when the data is > > exported via node exporter with collector.perf option enabled. > > Does the perf collector work at all for you? > What's the value of the kernel.perf_event_paranoid sysctl? > > I'm not seeing a reference to LLC here: > > > https://github.com/prometheus/node_exporter/blob/master/collector/perf_linux.go > > > Which could mean that this metric isn't supported yet, but might be easy > to add. > > Kind regards, > Christian > > > (Also, please don't cross-post :)) > On Tuesday, June 16, 2020 at 12:58:43 PM UTC+5:30, Christian Hoffmann wrote: > > Hi, > > On 6/16/20 7:42 AM, Divya Rajasekaran wrote: > > Is there any way to push 3rd party VNF metric into the node exporter > > Are you referring to Virtual Network Functions as described here? > https://tools.ietf.org/id/draft-rosa-bmwg-vnfbench-01.html > > I don't think node_exporter has support for this. > > Could this be what you are looking for? > https://osm.etsi.org/wikipub/index.php/OSM_Performance_Management > > If this doesn't help, you might have to look for another solution. If > this is only about some additional machine-specific metrics, it might be > a good fit for a pull request for node_exporter (check with the > maintainers first). A simpler approach would be using the > node_exporter's textfile collector for that and exporting the required > metrics to a "vnf.prom" file or something. > > If this is something bigger / non-machine specific, it might make sense > to use separate exporter. > > > and also I am not able to see LLC-stores metrics when the data is > > exported via node exporter with collector.perf option enabled. > > Does the perf collector work at all for you? > What's the value of the kernel.perf_event_paranoid sysctl? > > I'm not seeing a reference to LLC here: > > > https://github.com/prometheus/node_exporter/blob/master/collector/perf_linux.go > > > Which could mean that this metric isn't supported yet, but might be easy > to add. > > Kind regards, > Christian > > > (Also, please don't cross-post :)) > On Tuesday, June 16, 2020 at 12:58:43 PM UTC+5:30, Christian Hoffmann wrote: > > Hi, > > On 6/16/20 7:42 AM, Divya Rajasekaran wrote: > > Is there any way to push 3rd party VNF metric into the node exporter > > Are you referring to Virtual Network Functions as described here? > https://tools.ietf.org/id/draft-rosa-bmwg-vnfbench-01.html > > I don't think node_exporter has support for this. > > Could this be what you are looking for? > https://osm.etsi.org/wikipub/index.php/OSM_Performance_Management > > If this doesn't help, you might have to look for another solution. If > this is only about some additional machine-specific metrics, it might be > a good fit for a pull request for node_exporter (check with the > maintainers first). A simpler approach would be using the > node_exporter's textfile collector for that and exporting the required > metrics to a "vnf.prom" file or something. > > If this is something bigger / non-machine specific, it might make sense > to use separate exporter. > > > and also I am not able to see LLC-stores metrics when the data is > > exported via node exporter with collector.perf option enabled. > > Does the perf collector work at all for you? > What's the value of the kernel.perf_event_paranoid sysctl? > > I'm not seeing a reference to LLC here: > > > https://github.com/prometheus/node_exporter/blob/master/collector/perf_linux.go > > > Which could mean that this metric isn't supported yet, but might be easy > to add. > > Kind regards, > Christian > > > (Also, please don't cross-post :)) > On Tuesday, June 16, 2020 at 12:58:43 PM UTC+5:30, Christian Hoffmann wrote: > > Hi, > > On 6/16/20 7:42 AM, Divya Rajasekaran wrote: > > Is there any way to push 3rd party VNF metric into the node exporter > > Are you referring to Virtual Network Functions as described here? > https://tools.ietf.org/id/draft-rosa-bmwg-vnfbench-01.html > > I don't think node_exporter has support for this. > > Could this be what you are looking for? > https://osm.etsi.org/wikipub/index.php/OSM_Performance_Management > > If this doesn't help, you might have to look for another solution. If > this is only about some additional machine-specific metrics, it might be > a good fit for a pull request for node_exporter (check with the > maintainers first). A simpler approach would be using the > node_exporter's textfile collector for that and exporting the required > metrics to a "vnf.prom" file or something. > > If this is something bigger / non-machine specific, it might make sense > to use separate exporter. > > > and also I am not able to see LLC-stores metrics when the data is > > exported via node exporter with collector.perf option enabled. > > Does the perf collector work at all for you? > What's the value of the kernel.perf_event_paranoid sysctl? > > I'm not seeing a reference to LLC here: > > > https://github.com/prometheus/node_exporter/blob/master/collector/perf_linux.go > > > Which could mean that this metric isn't supported yet, but might be easy > to add. > > Kind regards, > Christian > > > (Also, please don't cross-post :)) > -- You received this message because you are subscribed to the Google Groups "Prometheus Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/f541c70e-b4a7-4d3d-bfcd-4f4b0529c36fo%40googlegroups.com.
