Bug 1883206 - rgw: prioritize swift healthcheck w/mclock qos
Summary: rgw: prioritize swift healthcheck w/mclock qos
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Red Hat Ceph Storage
Classification: Red Hat Storage
Component: RGW
Version: 4.1
Hardware: Unspecified
OS: Unspecified
low
medium
Target Milestone: ---
: 8.0
Assignee: Mark Kogan
QA Contact: Madhavi Kasturi
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2020-09-28 12:21 UTC by Matt Benjamin (redhat)
Modified: 2023-08-17 12:46 UTC (History)
7 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2023-08-16 15:06:43 UTC
Embargoed:


Attachments (Terms of Use)

Description Matt Benjamin (redhat) 2020-09-28 12:21:42 UTC
Description of problem:

A recent rook issue report suggests that under various circumstances (dynamic bucket index reshard event) the swift healthcheck can be delayed, causing liveness checks to fail.

https://github.com/rook/rook/issues/6304

Since the swift healthcheck has minimal overhead and generates no cluster traffic, this is unexpected, but may be due to lack of request prioritization.

Comment 1 RHEL Program Management 2020-09-28 12:21:50 UTC
Please specify the severity of this bug. Severity is defined here:
https://bugzilla.redhat.com/page.cgi?id=fields.html#bug_severity.

Comment 2 Mark Kogan 2020-10-14 14:20:29 UTC
In testing on a vstart.sh cluster (HDDs) while resharding 10M objects bucket
checking the watchdog http GET operation latency to the /swift/healthcheck url via curl:
  curl -sv -D - "http://127.0.0.1:8000/swift/healthcheck"  
the latency is not affected by ongoing resharding.

Added a reequest for more information to the github issue.


Note You need to log in before you can comment on or make changes to this bug.