Project

General

Profile

Bug #11792

mds: recursive statistics are either inaccurate or too "chunky"

Added by Greg Farnum over 3 years ago. Updated 8 months ago.

Status:
New
Priority:
High
Assignee:
-
Category:
Administration/Usability
Target version:
Start date:
05/27/2015
Due date:
% Done:

0%

Source:
Community (user)
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(FS):
Client, MDS, kceph
Labels (FS):
task(medium)
Pull request ID:

Description

https://www.mail-archive.com/ceph-users@lists.ceph.com/msg20005.html

Thanks Wido for this very interesting (and very simple) feature.
But does it work well? Because, I use Hammer in a Ubuntu Trusty
cluster nodes, and in a Ubuntu Trusty client with 3.16 kernel
and cephfs mounted with the kernel module client, I have this:

~# mount | grep cephfs # /mnt is my mounted cephfs
10.0.2.150,10.0.2.151,10.0.2.152:/ on /mnt type ceph 
(noacl,name=cephfs,key=client.cephfs)

~# ls -lah /mnt/dir1/
total 0
drwxr-xr-x 1 root root  96M May 12 21:06 .
drwxr-xr-x 1 root root 103M May 17 23:56 ..
drwxr-xr-x 1 root root  96M May 12 21:06 8
drwxr-xr-x 1 root root 4.0M May 17 23:57 test

As you can see:
  /mnt/dir1/8/      => 96M
  /mnt/dir1/test/   => 4.0M

But:
  /mnt/dir1/ (ie .) => 96M

I should have:

    size("/mnt/dir1/") = size("/mnt/dir1/8/") + size("/mnt/dir1/test/")

I have a suspicion this is a result of changing (one of?) our block sizes to 4MB for the stat results, but I'm not sure if it's being misinterpreted by this particular stack or if we're now stuck on all stacks reporting stuff in 4MB multiples. (For every file, or just for directories?)

History

#1 Updated by Zheng Yan over 3 years ago

thanks to delay propagation, recursive statistics are never accurate

#2 Updated by Greg Farnum over 3 years ago

Well, they're delayed, yes. But the few times I've bothered to measure the delay is very short (a few seconds at most) and the sizes reported (4MB and 96MB, combining for 96MB) are making me very suspicious.

#3 Updated by Greg Farnum over 2 years ago

  • Priority changed from Normal to High

This is going to pop up for users. Let's be proactive about checking and fixing it if possible.

#4 Updated by Greg Farnum over 2 years ago

  • Component(FS) kceph added

#5 Updated by Greg Farnum over 2 years ago

  • Category changed from 53 to Administration/Usability

#6 Updated by Patrick Donnelly 8 months ago

  • Subject changed from recursive statistics are either inaccurate or too "chunky" to mds: recursive statistics are either inaccurate or too "chunky"
  • Target version set to v14.0.0
  • Component(FS) Client, MDS added
  • Labels (FS) task(medium) added

Also available in: Atom PDF