Project

General

Profile

Actions

Feature #4386

closed

kclient: Mount error message when no MDS present

Added by Mark Nelson about 11 years ago. Updated over 4 years ago.

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

0%

Source:
Development
Tags:
Backport:
Reviewed:
Affected Versions:
Component(FS):
kceph
Labels (FS):
Pull request ID:

Description

Right now you either get an input/output error or a message about not being able to find the superblock when trying to mount cephfs when there is no MDS. This is confusing for new users as they have no idea what's going on. We should probably report on the command line and in the logs when a user is trying to mount CephFS with no MDS present.

Actions #1

Updated by Greg Farnum almost 9 years ago

  • Subject changed from Mount error message when no MDS present to kclient: Mount error message when no MDS present
  • Category set to 53
Actions #2

Updated by Greg Farnum almost 8 years ago

  • Component(FS) kceph added
Actions #3

Updated by Greg Farnum almost 8 years ago

  • Category changed from 53 to Administration/Usability
Actions #4

Updated by Patrick Donnelly over 4 years ago

  • Assignee set to Xiubo Li
  • Start date deleted (03/07/2013)
Actions #5

Updated by Xiubo Li over 4 years ago

  • Status changed from New to In Progress

Currently from my test this has been fixed by e9e427f0a14f7.
Will go through the related code and test it more to make sure all this works as expected.

Thanks

Actions #6

Updated by Xiubo Li over 4 years ago

An extra patch has been post and it is based on the current old mount API.
There is a new mount API for cephfs, and I will test it and to see whether should we fix it like this later.

Actions #7

Updated by Xiubo Li over 4 years ago

Checked the new mount API, we still need the fix when the mount request timedout due to there is no any MDS is up or the cluster is laggy.

Actions #8

Updated by Xiubo Li over 4 years ago

And maybe we could return the -ESTALE or some other specified errornos to the userland to mount.ceph and then the mount.ceph could print some hint message about "no MDS is up or cluster is laggy".

Actions #9

Updated by Xiubo Li over 4 years ago

Return -EHOSTUNREACH instead if no MDS is up or the cluster is laggy.

Actions #10

Updated by Xiubo Li over 4 years ago

  • Status changed from In Progress to Fix Under Review
Actions #12

Updated by Xiubo Li over 4 years ago

  • Status changed from Fix Under Review to Resolved
Actions

Also available in: Atom PDF