Previously, the metrics charts on the node page could incorrectly total usage for more than one node in some circumstances. The node page charts now correctly display the usage only for that node.
Description of problem:
Metrics on node overview page are defined following way (e.g. for memory):
node_memory_Active_bytes{instance=~'.*10.0.149.13.*'}
Problem is in '.*10.0.149.13.*'
We have two nodes named:
host-10-0-149-13
host-10-0-149-136
with IPs:
10.0.149.13
10.0.149.136
So the '.*10.0.149.13.*' selector matches both nodes and UI console shows on node overview page for host-10-0-149-13 incorrect metrics which are relevant for other host host-10-0-149-136.
Version-Release number of selected component (if applicable):
4.1.0
Kubernetes Master Version
v1.13.4+838b4fa
How reproducible:
Always
Steps to Reproduce:
1. have two nodes named as described above with described IPs (not sure if names or IPs are relevant for the {instance=~'.*10.0.149.13.*'} expr)
2. go to node overview in UI console for host-10-0-149-13
Actual results:
node overview page for host-10-0-149-13 shows incorrect metrics which are relevant for other host host-10-0-149-136.
Expected results:
Metrics for host-10-0-149-13 node should be shown.
The node selector is' ${nodeIp.address}:.*'
click one metrics such as Memory Usage on web console, it will got selector like node_memory_Active_bytes{instance=~'10.0.142.139:.*'}
Verify this bug.
4.2.0-0.nightly-2019-06-24-160709
quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:24d772dd515d16c28d13de6c349659df0d9807aabb512a25b49d51df82de26d2
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.
For information on the advisory, and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.
https://access.redhat.com/errata/RHBA-2019:2922
Description of problem: Metrics on node overview page are defined following way (e.g. for memory): node_memory_Active_bytes{instance=~'.*10.0.149.13.*'} Problem is in '.*10.0.149.13.*' We have two nodes named: host-10-0-149-13 host-10-0-149-136 with IPs: 10.0.149.13 10.0.149.136 So the '.*10.0.149.13.*' selector matches both nodes and UI console shows on node overview page for host-10-0-149-13 incorrect metrics which are relevant for other host host-10-0-149-136. Version-Release number of selected component (if applicable): 4.1.0 Kubernetes Master Version v1.13.4+838b4fa How reproducible: Always Steps to Reproduce: 1. have two nodes named as described above with described IPs (not sure if names or IPs are relevant for the {instance=~'.*10.0.149.13.*'} expr) 2. go to node overview in UI console for host-10-0-149-13 Actual results: node overview page for host-10-0-149-13 shows incorrect metrics which are relevant for other host host-10-0-149-136. Expected results: Metrics for host-10-0-149-13 node should be shown.