Bug 1003802 - Regression Multihost/idmapping test fail
Summary: Regression Multihost/idmapping test fail
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: Red Hat Enterprise Linux 6
Classification: Red Hat
Component: nfs-utils
Version: 6.5
Hardware: All
OS: Linux
unspecified
unspecified
Target Milestone: rc
: ---
Assignee: Steve Dickson
QA Contact: Red Hat Kernel QE team
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2013-09-03 09:16 UTC by JianHong Yin
Modified: 2018-12-06 15:16 UTC (History)
2 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2013-11-08 03:23:51 UTC
Target Upstream Version:


Attachments (Terms of Use)

Description JianHong Yin 2013-09-03 09:16:40 UTC
Description of problem:
Regression Multihost/idmapping test fail

Version-Release number of selected component (if applicable):
RHEL-6.5-20130830.2

How reproducible:
always

Steps to Reproduce:
1. run /CoreOS/nfs-utils/Multihost/idmapping
2.
3.

Actual results:
fail

Expected results:
pass

Additional info:

Comment 1 JianHong Yin 2013-09-03 09:17:20 UTC
restart rpcidmapd fail:

::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
:: [   LOG    ] :: Client/Setup
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

:: [   PASS   ] :: Add a new user _qa (Expected 0, got 0)
:: [   PASS   ] :: Add a new user __qa (Expected 0, got 0)
:: [   PASS   ] :: Add a new user qa (Expected 0, got 0)
:: [ 04:52:41 ] :: [ INFO    ] :: Domain = test.com
Shutting down RPC idmapd: [FAILED]
Starting RPC idmapd: rpc.idmapd: libnfsidmap: using domain: test.com
rpc.idmapd: libnfsidmap: Realms list: 'TEST.COM' 
rpc.idmapd: libnfsidmap: processing 'Method' list
rpc.idmapd: libnfsidmap: loaded plugin /usr/lib64/libnfsidmap/nsswitch.so for method nsswitch

[FAILED]
:: [   FAIL   ] :: Restart rpcidmapd service (Expected 0, got 1)
:: [   PASS   ] :: Create nfs mount point was successful (Expected 0, got 0)

Comment 3 JianHong Yin 2013-09-04 06:02:53 UTC
(In reply to Yin.JianHong from comment #1)
more info:
test in RHEL-6.1 PASS
   https://beaker.engineering.redhat.com/jobs/487452
test in RHEL-6.2 PASS
   https://beaker.engineering.redhat.com/jobs/487451
test in RHEL-6.3 FAIL
   https://beaker.engineering.redhat.com/jobs/487450
test in RHEL-6.4 FAIL
   https://beaker.engineering.redhat.com/jobs/487404
test in RHEL-6.5-2013-0830.2  FAIL
   https://beaker.engineering.redhat.com/jobs/487406

Comment 4 JianHong Yin 2013-09-04 08:23:02 UTC
https://beaker.engineering.redhat.com/jobs/487553

:: [   PASS   ] :: touch /nfs-mount/file3 (Expected 0, got 0)
-rw-r--r--. 1 root root 0 Sep  4 03:47 /nfs-mount/file3
root.root
:: [   PASS   ] :: Test file3 has correct uid.gid(root.root) (Expected 0, got 0)
:: [   PASS   ] :: Test root could chown file3 to qa.qa (Expected 0, got 0)
-rw-r--r--. 1 nobody nobody 0 Sep  4 03:47 /nfs-mount/file3
nobody.nobody
:: [   FAIL   ] :: Test file3 uid.gid change to qa.qa  (Expected 0, got 1)
:: [ 03:47:17 ] ::  
===> /var/log/messages:
Sep  4 03:47:14 dell-pe650-02 nfsidmap[2704]: key: 0x6617212 type: uid value: nobody timeout 600
Sep  4 03:47:14 dell-pe650-02 nfsidmap[2704]: nss_getpwnam: name 'nobody' does not map into domain 'test.com'
Sep  4 03:47:14 dell-pe650-02 nfsidmap[2706]: key: 0x2f35fc2d type: gid value: nobody timeout 600


chown return success, but ls & stat display the gid/uid nobody

Comment 5 Steve Dickson 2013-09-04 13:16:09 UTC
(In reply to Yin.JianHong from comment #4)
> https://beaker.engineering.redhat.com/jobs/487553
> 
> :: [   PASS   ] :: touch /nfs-mount/file3 (Expected 0, got 0)
> -rw-r--r--. 1 root root 0 Sep  4 03:47 /nfs-mount/file3
> root.root
> :: [   PASS   ] :: Test file3 has correct uid.gid(root.root) (Expected 0,
> got 0)
> :: [   PASS   ] :: Test root could chown file3 to qa.qa (Expected 0, got 0)
> -rw-r--r--. 1 nobody nobody 0 Sep  4 03:47 /nfs-mount/file3
> nobody.nobody
> :: [   FAIL   ] :: Test file3 uid.gid change to qa.qa  (Expected 0, got 1)
> :: [ 03:47:17 ] ::  
> ===> /var/log/messages:
> Sep  4 03:47:14 dell-pe650-02 nfsidmap[2704]: key: 0x6617212 type: uid
> value: nobody timeout 600
> Sep  4 03:47:14 dell-pe650-02 nfsidmap[2704]: nss_getpwnam: name 'nobody'
> does not map into domain 'test.com'
Would it be possible to get a network trace to see what the
server is sending over the wire?

Also what exactly is this testing doing?

Comment 6 JianHong Yin 2013-09-04 23:03:28 UTC
(In reply to Steve Dickson from comment #5)
> (In reply to Yin.JianHong from comment #4)
> > https://beaker.engineering.redhat.com/jobs/487553
> > 
> > :: [   PASS   ] :: touch /nfs-mount/file3 (Expected 0, got 0)
> > -rw-r--r--. 1 root root 0 Sep  4 03:47 /nfs-mount/file3
> > root.root
> > :: [   PASS   ] :: Test file3 has correct uid.gid(root.root) (Expected 0,
> > got 0)
> > :: [   PASS   ] :: Test root could chown file3 to qa.qa (Expected 0, got 0)
> > -rw-r--r--. 1 nobody nobody 0 Sep  4 03:47 /nfs-mount/file3
> > nobody.nobody
> > :: [   FAIL   ] :: Test file3 uid.gid change to qa.qa  (Expected 0, got 1)
> > :: [ 03:47:17 ] ::  
> > ===> /var/log/messages:
> > Sep  4 03:47:14 dell-pe650-02 nfsidmap[2704]: key: 0x6617212 type: uid
> > value: nobody timeout 600
> > Sep  4 03:47:14 dell-pe650-02 nfsidmap[2704]: nss_getpwnam: name 'nobody'
> > does not map into domain 'test.com'
> Would it be possible to get a network trace to see what the
> server is sending over the wire?
I will try to get network trace later, and reply,,,

> 
> Also what exactly is this testing doing?
In the Failed section, do the following test:
1. As root, touch a file in /nfs-mount, and check the file's uid/gid
   ls -l;  stat;
   Expected results: root.root
2. chown the file to qa:qa, check if chown success
   check exit code;
   Expected results: 0
3. and check the uid/gid   <<<-- FAIL in this step, in RHEL6.1 RHEL6.2 PASS
   ls -l; stat
   Expected results: qa.qa
   Actual results: nobody.nobody

Comment 7 RHEL Program Management 2013-10-14 02:24:54 UTC
This request was not resolved in time for the current release.
Red Hat invites you to ask your support representative to
propose this request, if still desired, for consideration in
the next release of Red Hat Enterprise Linux.

Comment 11 JianHong Yin 2013-11-08 03:23:51 UTC
Now I understand this is a feather implementation from bz705099,
but turn on/off by modules parameter is not a good UX.

Because this a feather of idmapd service,
Suggest add an option for rpc.idmapd, and user can choose turn on or off by
RPCIDMAPDARGS= in /etc/sysconfig/nfs.

I will close this, and open an other RFE type bugzilla:
https://bugzilla.redhat.com/show_bug.cgi?id=1028269


Note You need to log in before you can comment on or make changes to this bug.