Default rules not working correctly

[fajfer]

With the defaults now as they are CPU,RAM and DiskSpace alerts needs to be reworked.

eg.
(100 * (1 - avg(irate(node_cpu{job="node",mode="idle"}[5m])) BY (instance))) > 96
won't work, but
100 - (avg by (instance) (irate(node_cpu_seconds_total{job="node",mode="idle"}[5m])) * 100) > 96
will

I think the PromQL just changed as I remember those defaults to be working correctly (unless I am missing something and the first one is ok).

[paulfantom]

@SuperQ could you look at those expressions?

[fajfer]

I've checked the rest just now.
CriticalRAMUsage is:
(1 - ((node_memory_MemFree + node_memory_Buffers + node_memory_Cached) / node_memory_MemTotal)) * 100 > 98
should be:
(1 - ((node_memory_MemFree_bytes + node_memory_Buffers_bytes + node_memory_Cached_bytes) / node_memory_MemTotal_bytes)) * 100 > 98

CriticalDiskSpace is:
node_filesystem_free{job="node",filesystem!~"^/run(/|$)"} / node_filesystem_size{job="node"} < 0.1
should be:
node_filesystem_free_bytes{job="node",filesystem!~"^/run(/|$)"} / node_filesystem_size_bytes{job="node"} < 0.1

[SuperQ]

The expressions are fine, but the metric names have changed.

[fajfer]

The expressions are fine, but the metric names have changed.

That's the point

[lock]

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.