Project

Profile

Help

Issue #915

/status/ API does not report health of webservers

Added by bmbouter over 4 years ago. Updated 6 months ago.

Status:
CLOSED - WONTFIX
Priority:
Normal
Assignee:
-
Category:
-
Sprint/Milestone:
-
Start date:
Due date:
Severity:
1. Low
Version:
Platform Release:
Blocks Release:
OS:
Backwards Incompatible:
No
Triaged:
Yes
Groomed:
No
Sprint Candidate:
No
Tags:
Pulp 2
QA Contact:
Complexity:
Smash Test:
Verified:
No
Verification Required:
No
Sprint:

Description

When using the /status/ API to evaluate the health of a Pulp system, webserver availability and health information is not included. In a standalone installation this is not needed because the response code of 200 OK is intended to indicate the webserver is OK. In a clustered environment with multiple webservers, a /status/ request may be handled by a webserver who is online and functioning, while other webservers may be offline.

To workaround this limitation, it is recommended that all webservers in a cluster have the /status/ API call monitored directly (not through the load balancer). All webservers that are expected to be online should be checked that they return a '200 OK'.

This is likely not something Pulp can resolve currently because the load balancer is the component that knows which webservers to expect to be online or not. I propose this bug remain open to track this issue as a known limitation.

Associated revisions

Revision e00e8d5a View on GitHub
Added by bmbouter over 4 years ago

Adds three updates to the clustering documentation.

  • A section on clustered monitoring which identifies issues and
    workarounds for the /status/ API in clustered environments.
  • A section on pulp-admin configuration for clustered
    environments.
  • Generalizes the load balancer docs to include DNS based load
    balancing in addition to TCP based load balancing.

re #915
re #282

https://pulp.plan.io/issues/915
https://pulp.plan.io/issues/282

Revision e00e8d5a View on GitHub
Added by bmbouter over 4 years ago

Adds three updates to the clustering documentation.

  • A section on clustered monitoring which identifies issues and
    workarounds for the /status/ API in clustered environments.
  • A section on pulp-admin configuration for clustered
    environments.
  • Generalizes the load balancer docs to include DNS based load
    balancing in addition to TCP based load balancing.

re #915
re #282

https://pulp.plan.io/issues/915
https://pulp.plan.io/issues/282

Revision e00e8d5a View on GitHub
Added by bmbouter over 4 years ago

Adds three updates to the clustering documentation.

  • A section on clustered monitoring which identifies issues and
    workarounds for the /status/ API in clustered environments.
  • A section on pulp-admin configuration for clustered
    environments.
  • Generalizes the load balancer docs to include DNS based load
    balancing in addition to TCP based load balancing.

re #915
re #282

https://pulp.plan.io/issues/915
https://pulp.plan.io/issues/282

History

#1 Updated by bmbouter over 4 years ago

  • Description updated (diff)

#2 Updated by cduryee over 4 years ago

  • Status changed from NEW to CLOSED - WONTFIX
  • Triaged changed from No to Yes

marking issue as closed/wontfix. The workaround is documented in referenced commit.

#3 Updated by bmbouter 6 months ago

  • Tags Pulp 2 added

Please register to edit this issue

Also available in: Atom PDF