Project

General

Profile

Actions
Copy link

Bug #36746

open

Ignore osd_find_best_info_ignore_history_les for erasure-coded PGs

Added by David Zafman over 5 years ago. Updated over 4 years ago.

Status:
New
Priority:
Normal
Assignee:
David Zafman
Category:
-
Target version:
-
% Done:

0%

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

Description

The only case that osd_find_best_info_ignore_history_les would work for erasure coded pools is if an interval didn't have changes anyway. If a shard is allowed to peer for an erasure coded pool it can only cause incorrectly decoded objects if needing to recover/backfill. I believe this is what caused a crash:

/builddir/build/BUILD/ceph-12.2.5/src/osd/ECUtil.cc: 59: FAILED assert(i->second.length() == total_data_size)

Actions
Copy link
 #1

Updated by Greg Farnum over 5 years ago

  • Project changed from Ceph to RADOS
  • Category deleted (OSD)
Actions
Copy link
 #2

Updated by David Zafman almost 5 years ago

  • Assignee set to David Zafman
Actions
Copy link
 #3

Updated by David Zafman over 4 years ago

Maybe change this from true/false to specify a PG, so only that PG is affected.

Actions
Copy link
 #4

Updated by Josh Durgin over 4 years ago

  • Priority changed from High to Normal
Actions
Copy link

Also available in: Atom PDF