Sorry but where to add all of these modifications ? the node_exporter file on my remote host is a binary file ?!
On Wednesday, June 17, 2020 at 7:31:41 PM UTC+1, Ben Kochie wrote: > > Yes, we use a recording rules to represent node memory utilization. > > > https://gitlab.com/gitlab-com/runbooks/-/blob/1682a5632f3eaf0548dfa8277a421de2aff24245/rules/node.yml#L91-106 > > On Wed, Jun 17, 2020 at 6:14 PM Christian Hoffmann < > [email protected] <javascript:>> wrote: > >> Hi, >> >> On 6/17/20 4:44 PM, Yasmine Mbarek wrote: >> > I have a tiny problem with node exporter. If you can help me I will be >> > very grateful . >> > So my node exporter implemented in my parc of machines , for some >> > machine it works fine and returns all metrics values but in other >> > machine it returns everything but "RAM Used" , the difference between >> > the working machines and the others is that : on the machines that RAM >> > used is working fine , the OS is redhat 7 >> > the rest of machines Redhat 6 everything is working but the RAM used >> > returns NO DATA >> > Is there any explanation ?? >> >> I assume you are using Grafana with some kind of node_exporter dashboard? >> >> You would have to look into the actual queries to see what's causing this. >> >> My guess: The dashboard most likely uses the MemAvailable metric (which >> comes from /proc/meminfo). This is exposed by the Linux kernel, but only >> after some specific version. RHEL6 does not expose it. >> >> There are ways to calculate something similar for RHEL6 (the "free" >> command line tool has some logic for this). >> >> I think GitLab had a public Prometheus config with a recording rule, >> maybe it was this one: >> >> >> https://gitlab.com/gitlab-org/gitlab-foss/-/commit/e91c7469ad0be5f429548d4142ca93c17ec9e71e >> >> You would have to set up such a recording rule and would have to modify >> your dashboard accordingly. >> >> As an alternative: Try talking your administrators into abandoning >> RHEL6. It'll be out of support at the end of the year anyway, IIRC. :) >> >> Kind regards, >> Christian >> >> -- >> 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] <javascript:>. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/prometheus-users/6fbaa443-48f3-1bd0-962d-490e4cde4b80%40hoffmann-christian.info >> . >> > -- 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/92930954-4ebb-4af7-b739-c9d63ae1cd44o%40googlegroups.com.
