Loading...

XML

Word

Printable

Type: Story
Resolution: Unresolved
Priority: Major
Fix Version/s: None
Affects Version/s: None
Component/s: None
Labels:
- groomed

Story Points:
3
Blocked:
False
Ready:
False
Epic Link:
Production/public instance of OSUS should be able to scale without causing issues in a multi-tenant environment- phase2
Market:

Sprint:
OTA 256, OTA 257, OTA 258, OTA 259, OTA 260, OTA 264

Target Version:

openshift-4.18

SFDC Cases Links:
SFDC Cases Counter:
SFDC Cases Open:

Currently we manually bump amount of instances depending on amount of incoming requests. We should add HPA based on
irate(cincinnati_pe_graph_incoming_requests_total[1m])
query, so that the autoscaler would create additional Cincy instances when load is high

is blocked by

OTA-869 move graph-builder to separate pod

To Do

is related to

OTA-1372 Add keda support to ota-stage cluster

Closed

relates to

OTA-1083 Adjust cincinnati replicas after root cause analyzed

In Progress

links to

openshift/cincinnati#937: OTA-543: add autoscaler to cincinnati deployment

openshift/cincinnati#942: OTA-543: Add PrometheusRule to create a recording rule for calculating avg requests

openshift/cincinnati#959: OTA-543: fix the PE_REQ_AVG templating

openshift/cincinnati#973: OTA-543: remove prometheus rule and add query directly to scaledObject

mentioned on

Merge request - adjust min and max replica count for cincinnati deployments

Merge request - bump cincinnati to e9323171 and start using autoscalers

Merge request - cincinnati to manage HPA and remove DC

Merge request - Install custom metrics autoscaler to ota-stage cluster

Merge request - temporarily add hpa to cleanup resources

Merge request - use scaledobject instead of HPA in cincinnati

(2 links to, 6 mentioned on)

Assignee:: Pratik Mahajan

Reporter:: Vadim Rutkovsky

Votes:: 0 Vote for this issue

Watchers:: 6 Start watching this issue

Created:: 2021/12/14 1:51 PM

Updated:: 2024/11/04 8:32 PM