Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.
This project is now read‑only. Starting Monday, February 2, please use https://ibm-ceph.atlassian.net/ for all bug tracking management.

Bug 1526446

Summary: [RFE] Enhance RGW Garbage Collection Capabilities to Mitigate Client Contention
Product: [Red Hat Storage] Red Hat Ceph Storage Reporter: Benjamin Schmaus <bschmaus>
Component: RGWAssignee: J. Eric Ivancich <ivancich>
Status: CLOSED DUPLICATE QA Contact: ceph-qe-bugs <ceph-qe-bugs>
Severity: medium Docs Contact:
Priority: unspecified    
Version: 2.4CC: cbodley, ceph-eng-bugs, kbader, mbenjamin, mhackett, owasserm, sweil
Target Milestone: rcKeywords: FutureFeature
Target Release: 3.1   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2018-02-27 17:38:25 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Benjamin Schmaus 2017-12-15 14:00:45 UTC 
Description of problem:
On a busy RGW gateway customer experiences issues where GC will pre-empt client RGW requests because the GC is busy doing cleanup.

Need to look at multiple options to fix:

1) Allow GC to be configurable on certain RGWs so that client requests can go to some RGWs and the GC runs on its own RGW host with no client requests thus mitigating the resource contention

2) Put some type of QOS policy in place for GC so that it never impacts client requests on a RGW

3) Provide ability to throttle GC requests

4) Some other approach not noted above


Version-Release number of selected component (if applicable):
2.4

How reproducible:
Sporadic but on a busy cluster usually regularly

Steps to Reproduce:
1.Create a busy cluster with lots of objects being written and deleted
2.GC will start to perform its tasks 
3.Clients will experience slow downs

Actual results:
Pre-empted client requests

Expected results:
Client requests should remain unimpaired

Additional info:
Comment 4 Benjamin Schmaus 2018-02-27 17:38:25 UTC 

*** This bug has been marked as a duplicate of bug 1548564 ***