Bug 763729 (GLUSTER-1997) - Solaris mount fails with "RPC program not registered"
Summary: Solaris mount fails with "RPC program not registered"
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: GLUSTER-1997
Product: GlusterFS
Classification: Community
Component: nfs
Version: 3.1.0
Hardware: All
OS: Linux
low
medium
Target Milestone: ---
Assignee: Shehjar Tikoo
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2010-10-21 09:05 UTC by Shehjar Tikoo
Modified: 2015-12-01 16:45 UTC (History)
2 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed:
Regression: RTP
Mount Type: nfs
Documentation: DA
CRM:
Verified Versions:

Attachments (Terms of Use)
Add an attachment (proposed patch, testcase, etc.)

Description Shehjar Tikoo 2010-10-21 06:41:52 UTC 
The problem was occurring because the solaris root fh funging code wasnt tested for dvm against a solaris client.
Comment 1 Shehjar Tikoo 2010-10-21 09:05:24 UTC 
bash-3.00# showmount -e localhost    
export list for localhost:
/new *
bash-3.00# mount -o proto=tcp,vers=3 nfs://localhost:38467:/new  /mnt/gluster
nfs mount: localhost: : RPC: Program not registered
nfs mount: retrying: /mnt/gluster

Network trace shows that the Solaris client is receiving a ESTALE for lookup on root. The log shows this as:


[2010-10-21 08:38:04.895582] D [rpcsvc.c:1842:nfs_rpcsvc_request_create] nfsrpc: RPC XID: 4cb25ebf, Ver: 2, Program: 100003, ProgVers: 3, Proc: 3
[2010-10-21 08:38:04.895601] T [auth-unix.c:67:nfs_auth_unix_authenticate] nfsrpc: Auth Info: machine name: Solaris-10, uid: 0, gid: 0
[2010-10-21 08:38:04.895614] D [rpcsvc.c:1285:nfs_rpcsvc_program_actor] nfsrpc: Actor found: NFS3 - LOOKUP
######################################################################################
######################################################################################
[2010-10-21 08:38:04.895682] D [nfs3-helpers.c:2254:nfs3_log_fh_entry_call] nfs-nfsv3: XID: 4cb25ebf, LOOKUP: args: FH: hashcount 0, exportid 00000000-0000-0000-0000-0000000000
00, gfid 00000000-0000-0000-0000-000000000000, name: new
[2010-10-21 08:38:04.895714] T [nfs-common.c:99:nfs_mntpath_to_xlator] nfs: Subvolume search: new
[2010-10-21 08:38:04.895733] D [nfs3-helpers.c:2374:nfs3_log_common_res] nfs-nfsv3: XID: 4cb25ebf, LOOKUP: NFS: 70(Invalid file handle), POSIX: 14(Bad address)
######################################################################################
######################################################################################
Comment 2 Anand Avati 2010-10-21 09:11:26 UTC 
PATCH: http://patches.gluster.com/patch/5547 in master (nfs3: Correctly funge solaris root lookup FH for DVM)
Comment 3 Shehjar Tikoo 2010-10-26 08:38:34 UTC 
The other crash seen as part of this bug was related to logging and that is fixed. For closure, solaris needs to be mounted with the following command:

mount -o proto=tcp,vers=3 nfs://<SERVER-ADDR>:38467/<EXPORT> <MNT-POINT>
Comment 4 Lakshmipathi G 2010-11-19 07:52:20 UTC 
verified with 3.1.1qa9
Comment 5 Shehjar Tikoo 2010-12-23 05:21:08 UTC 
Doc available as FAQ.
Note You need to log in before you can comment on or make changes to this bug.