fix the prom query to look for job node

[amitsagtani97]

While the rest of the dashboard uses job="node" selector to query metrics, this panel is still configured to use the job="node-exporter" selector, which break the panel.

Updated the panel query to keep the consensus over all the dashboard.

[flokli]

There's still some more references to job="node-exporter". Can you search and replace them too?

assets/kubernetes/rules.yaml
648:        node_cpu_seconds_total{mode="idle",job="node-exporter"}
655:        node_memory_MemAvailable_bytes{job="node-exporter"} or
657:          node_memory_Buffers_bytes{job="node-exporter"} +
658:          node_memory_Cached_bytes{job="node-exporter"} +
659:          node_memory_MemFree_bytes{job="node-exporter"} +
660:          node_memory_Slab_bytes{job="node-exporter"}
667:          rate(node_cpu_seconds_total{mode!="idle",mode!="iowait",mode!="steal",job="node-exporter"}[5m])

site/content/kubernetes/_index.md
2122:    node_cpu_seconds_total{mode="idle",job="node-exporter"}
2134:    node_memory_MemAvailable_bytes{job="node-exporter"} or
2136:      node_memory_Buffers_bytes{job="node-exporter"} +
2137:      node_memory_Cached_bytes{job="node-exporter"} +
2138:      node_memory_MemFree_bytes{job="node-exporter"} +
2139:      node_memory_Slab_bytes{job="node-exporter"}
2151:      rate(node_cpu_seconds_total{mode!="idle",mode!="iowait",mode!="steal",job="node-exporter"}[5m])