Project

General

Profile

Actions

Bug #22003

closed

[CephFS-Ganesha]MDS migrate will affect Ganesha service?

Added by Gemini Chen over 6 years ago. Updated about 5 years ago.

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

0%

Source:
Tags:
Backport:
Regression:
No
Severity:
2 - major
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(FS):
Client, Ganesha FSAL
Labels (FS):
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

[Ganesha Version]
Ganesha V2.4
630a35bef41aabf76f99532448d6154316a525e0

[Ceph Version]
ceph version 10.2.7 (50e863e0f4bc8f4b9e31156de690d765af245185)

1.upgrade from 0.94.5 to 10.2.7
2.fresh 10.2.7

[How to Use CephFS]
1.Client using ceph-fuse to mount the cephfs service,it looks like this: ceph-fuse m <xxx>:6789 <yyy>
2. Re-export CephFS mount point as Ganesha-NFS4 share point.
---------------------------------

EXPORT { # Export Id (mandatory, each EXPORT must have a unique Export_Id)
Export_Id = 77;

  1. Exported path (mandatory)
    #Path = /nonexistant;
    Path = /opt/ganesha_nfs;
  1. Pseudo Path (required for NFS v4)
    #Pseudo = /nonexistant;
    Pseudo = /opt/ganesha_nfs;
  1. Required for access (default is None)
  2. Could use CLIENT blocks instead
    Access_Type = RW;
Squash = No_Root_Squash;
  1. Exporting FSAL
    FSAL {
    Name = VFS;
    }
    }
    -----------------------------------
    3. Last use NFS4 share service.it looks like this: mount -t nfs4 <xxx>:<yyy> <zzz>

[Issue]
Now I meet a problem: when mds service do migrate action,client will meet IO issue like :"Stale file handle", IO interrupt .

[Debug]
I found there may some issue when mds service migrate under HA.

[Solution]
When IO application meet the above issue, then i will do the following steps:
0. Restart the Ganesha-NFS server service.
1. Stop the IO application
2. Enter the directory again
3. Then, it will work OK as we expected.

Actions

Also available in: Atom PDF