Project

General

Profile

Bug #50004

mon: Modify Paxos trim logic to be more efficient

Added by Aishwarya Mathuria almost 3 years ago. Updated almost 3 years ago.

Status:
Resolved
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
% Done:

0%

Source:
Tags:
Backport:
pacific, octopus, nautilus
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
rados
Component(RADOS):
Monitor
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

When the log ingest rate is higher than the trimming rate, monitors may end up consuming a lot of space and become unresponsive in the worst case. This is because currently the rate of trimming is bounded - maybe_trim() is called on every tick, where it trims "paxos_service_trim_max" at max.

The trimming logic needs to be smarter and change according to the log ingest rate.


Related issues

Copied to RADOS - Backport #50123: octopus: mon: Modify Paxos trim logic to be more efficient Resolved
Copied to RADOS - Backport #50124: pacific: mon: Modify Paxos trim logic to be more efficient Resolved
Copied to RADOS - Backport #50125: nautilus: mon: Modify Paxos trim logic to be more efficient Resolved

History

#1 Updated by Neha Ojha almost 3 years ago

  • Status changed from New to Fix Under Review
  • Backport set to pacific, octopus, nautilus

#2 Updated by Kefu Chai almost 3 years ago

  • Status changed from Fix Under Review to Pending Backport

#3 Updated by Backport Bot almost 3 years ago

  • Copied to Backport #50123: octopus: mon: Modify Paxos trim logic to be more efficient added

#4 Updated by Backport Bot almost 3 years ago

  • Copied to Backport #50124: pacific: mon: Modify Paxos trim logic to be more efficient added

#5 Updated by Backport Bot almost 3 years ago

  • Copied to Backport #50125: nautilus: mon: Modify Paxos trim logic to be more efficient added

#6 Updated by Loïc Dachary almost 3 years ago

  • Status changed from Pending Backport to Resolved

While running with --resolve-parent, the script "backport-create-issue" noticed that all backports of this issue are in status "Resolved" or "Rejected".

Also available in: Atom PDF