Oracle ZFS Storage Appliance : SMB Panic on Shadow Migration Share when Deleting or Modifying a File or Directory whilst Migration is in progress

Asset ID:	1-72-2231985.1
Update Date:	2017-02-20
Keywords:

Solution Type Problem Resolution Sure

Solution 2231985.1 : Oracle ZFS Storage Appliance : SMB Panic on Shadow Migration Share when Deleting or Modifying a File or Directory whilst Migration is in progress

Applies to:

Oracle ZFS Storage ZS5-4 - Version All Versions to All Versions [Release All Releases]
Oracle ZFS Storage ZS5-2 - Version All Versions to All Versions [Release All Releases]
Oracle ZFS Storage ZS4-4 - Version All Versions to All Versions [Release All Releases]
Oracle ZFS Storage ZS3-4 - Version All Versions to All Versions [Release All Releases]
Oracle ZFS Storage ZS3-2 - Version All Versions to All Versions [Release All Releases]
7000 Appliance OS (Fishworks)

Symptoms

When using Shadow copy to migrate the data to a ZFS Storage Appliance, the shares are mounted via SMB - if the user then moves or removes directories or shares, the ZFS Storage will panic after about 2-3 minutes.

Using NFS for the same shares has no effect and works as expected.

Cause

Here is an example of the panic string on the ZFS Storage :

panic[cpu32]/thread=ffffff05da95fc20: recursive mutex_enter, lp=fffff60f42274080 owner=ffffff05da95fc20 thread=ffffff05da95fc20

ffffff05da95f3d0 unix:mutex_exit_critical_size+24578 ()
ffffff05da95f430 unix:mutex_vector_enter+197 ()
ffffff05da95f460 smbsrv:smb_node_set_doc_ignore+23 ()
ffffff05da95f4c0 smbsrv:smb_fem_oplock_vnevent+74 ()
ffffff05da95f540 genunix:vhead_vnevent+c6 ()
ffffff05da95f5a0 genunix:fop_vnevent+3b ()
ffffff05da95f5d0 genunix:vnevent_remove+3d ()
ffffff05da95f6e0 zfs:zfs_remove+198 ()
ffffff05da95f750 genunix:fop_remove+106 ()
ffffff05da95f8a0 smbsrv:smb_vop_remove+75 ()
ffffff05da95f8f0 smbsrv:smb_fsop_remove+a8 ()
ffffff05da95f920 smbsrv:smb_node_execute_doc+4e ()
ffffff05da95f950 smbsrv:smb_node_dec_open_ofiles+4a ()
ffffff05da95f9a0 smbsrv:smb_ofile_close_open+142 ()
ffffff05da95f9e0 smbsrv:smb_ofile_close+59 ()
ffffff05da95fac0 smbsrv:smb2_close+7d ()
ffffff05da95fb00 smbsrv:smb2_dispatch_request_impl+237 ()
ffffff05da95fb20 smbsrv:smb2_dispatch_request+15 ()
ffffff05da95fb70 smbsrv:smb_conn_worker+88 ()
ffffff05da95fc00 genunix:taskq_d_thread+c1 ()
ffffff05da95fc10 unix:thread_start+8 ()

syncing file systems... done

Solution

RPE provided an IDR release that solves the issue.

Please contact Oracle Support to provide access to the IDR fix.

IDR is ak-nas-2013-06-05-6-12-1-2-3-3-1

References

<BUG:25462366> - RECURSIVE MUTEX_ENTER ON SMB SHARE WHILST STILL SHADOW MIGRATING SOURCE
<NOTE:166650.1> - Working Effectively With Oracle Support - Best Practices
<NOTE:1019887.1> - Sun Storage 7000 Unified Storage System: How to Collect a Support Bundle using the BUI or CLI

Attachments

This solution has no attachment