forked from zegelin/cassandra-exporter
-
Notifications
You must be signed in to change notification settings - Fork 49
Open
Description
Hi folks,
We are seeing a mismatch between what cassandra_os_recent_cpu_load_ratio and sar is reporting.
Here's the sar data for the same period:
03:40:01 AM CPU %user %nice %system %iowait %steal %idle
...
06:05:03 AM all 37.65 0.00 3.90 7.63 0.88 49.93
06:10:01 AM all 34.33 0.00 3.71 7.77 0.74 53.46
06:15:01 AM all 31.85 0.00 3.41 6.09 0.62 58.03
06:20:01 AM all 27.23 0.00 3.10 5.88 0.50 63.30
06:25:03 AM all 34.87 0.00 2.85 1.49 0.45 60.34
06:30:01 AM all 28.96 0.00 2.63 1.24 0.33 66.84
06:35:03 AM all 30.06 0.00 2.69 1.26 0.36 65.63
06:40:01 AM all 29.72 0.00 2.67 1.03 0.35 66.23
06:45:01 AM all 30.58 0.00 2.60 1.14 0.32 65.37
06:50:01 AM all 28.70 0.00 2.66 1.01 0.33 67.30
06:55:03 AM all 30.19 0.00 2.66 1.16 0.32 65.67
07:00:01 AM all 29.78 0.00 2.51 1.06 0.28 66.37
07:05:03 AM all 30.15 0.00 3.05 1.31 0.45 65.05
Is there a reason why the metrics are so different?
We're using Cassandra exporter 0.9.14.
Thanks,
Chris.
Metadata
Metadata
Assignees
Labels
No labels