Bug 682790

Summary: Support new Ethernet naming standards (from biosdevname)
Product: Red Hat Enterprise Linux 6 Reporter: Andrew Cathrow <acathrow>
Component: vdsmAssignee: Yotam Oron <yoron>
Status: CLOSED ERRATA QA Contact: Kiril Nesenko <knesenko>
Severity: high Docs Contact:
Priority: high    
Version: 6.1CC: abaron, bazulay, cpelland, danken, dfediuck, iheim, ilvovsky, rbalakri, rbuilta, ykaul
Target Milestone: rcKeywords: Reopened
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: vdsm-4.9-68.el6 Doc Type: Bug Fix
Doc Text:
Vdsm no longer assume that NICs are named eth*. This makes Vdsm compatible with the new (optional) device naming convention. https://fedoraproject.org/wiki/Features/ConsistentNetworkDeviceNaming
 Story Points: ---
Clone Of: Environment:
Last Closed: 2011-12-06 07:08:33 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
biosdevname writeup none

Description Andrew Cathrow 2011-03-07 15:42:32 UTC 
Created attachment 482712 [details]
biosdevname writeup

RHEL 6.1 is planned to add support for biosdevname (See bz's below and attachment).
To support this VDSM needs to be able to support the new naming scheme.
More details in the attached write up, but this means for example that we can't hardcode eth* names, the name length could be longer than the usual ethXY and also the parsing code needs to handle embedded "#" within the name.


Relevant RHEL BZs

DevNaming6.1 tracker
https://bugzilla.redhat.com/show_bug.cgi?id=501533

BiosdevnameBugs tracker
https://bugzilla.redhat.com/show_bug.cgi?id=682269
Comment 2 Dan Kenigsberg 2011-03-07 16:28:21 UTC 
Specifically, vdsm/utils.getHostUUID() and its twin vdsm_reg/deployUtils.getHostID() should stop assuming nic names are eth*.

At least the first may easily use netinfo.get().
Comment 3 RHEL Program Management 2011-04-04 01:55:57 UTC 
Since RHEL 6.1 External Beta has begun, and this bug remains
unresolved, it has been rejected as it is not proposed as
exception or blocker.

Red Hat invites you to ask your support representative to
propose this request, if appropriate and relevant, in the
next release of Red Hat Enterprise Linux.
Comment 5 Yotam Oron 2011-05-11 05:54:49 UTC 
Also moved NET_CONF_BACK_DIR to netinfo:
http://gerrit.usersys.redhat.com/#change,381
Comment 7 RHEL Program Management 2011-05-15 12:15:12 UTC 
Quality Engineering Management has reviewed and declined this request.  You may
appeal this decision by reopening this request.
Comment 13 Kiril Nesenko 2011-05-30 10:56:44 UTC 
Verified
ic119

need to restart vdsm
Comment 14 Andrew Cathrow 2011-08-30 14:57:54 UTC 
This has failed in the field during beta #1

Setting back to assigned and raising priority. 


> ##########################################################################################################################################
> 
> 2011-08-30 10:21:18,834 INFO [org.nogah.bll.VdsInstaller]
> (pool-11-thread-33) Installation of 10.x.x.x. Successfully connected
> to server ssh. (Stage:
> Connecting to Host)
> 2011-08-30 10:21:18,834 INFO [org.nogah.bll.VdsInstaller]
> (pool-11-thread-33) Installation of 10.x.x.x. Executing installation
> stage. (Stage: Get the
> unique vds id)
> 2011-08-30 10:21:18,834 INFO
> [org.nogah.utils.hostinstall.MinaInstallWrapper] (pool-11-thread-33)
> Invoking /bin/echo -e `/bin/bash -c /usr/sbin/dmidecode
> |/bin/awk ' /UUID/{ print $2; } ' | /usr/bin/tr '
> ' '_' && cat /sys/class/net/eth*/address |/bin/sort | /usr/bin/head
> --lines=1` on 10.x.x.x
> 2011-08-30 10:21:18,956 INFO [org.nogah.bll.VdsInstaller]
> (pool-11-thread-33) Installation of 10.x.x.x. Recieved message:
> 4C4C4544-0039-3610-8057-C4C
> 04F425131_
> . FYI. (Stage: Get the unique vds id)
> 2011-08-30 10:21:18,974 INFO [org.nogah.bll.VdsInstaller]
> (pool-11-thread-33) Installation of 10.x.x.x. Assigning unique id
> 4C4C4544-0039-3610-8057-C4C04F425131_ to Host. (Stage: Get the unique
> vds id)
> 2011-08-30 10:21:18,978 ERROR
> [org.nogah.utils.hostinstall.MinaInstallWrapper] (pool-11-thread-33)
> cat: /sys/class/net/eth*/address: No such file or directory
> 
> 2011-08-30 10:21:18,978 ERROR [org.nogah.bll.VdsInstaller]
> (pool-11-thread-33) Installation of 10.x.x.x. Error: cat:
> /sys/class/net/eth*/address: No such file or directory
> . (Stage: Upload Installation script to Host)
> 2011-08-30 10:21:18,978 INFO [org.nogah.bll.InstallerMessages]
> (pool-11-thread-33) VDS message: cat: /sys/class/net/eth*/address: No
> such file or directory
>
Comment 15 Russ Builta 2011-08-30 16:10:47 UTC 
(In reply to comment #14)

The hosts are rhel6.1 vm's on Dell PowerEdge R810's - with Xeon E7540 processors HW.


> This has failed in the field during beta #1
> 
> Setting back to assigned and raising priority. 
> 
> 
> > ##########################################################################################################################################
> > 
> > 2011-08-30 10:21:18,834 INFO [org.nogah.bll.VdsInstaller]
> > (pool-11-thread-33) Installation of 10.x.x.x. Successfully connected
> > to server ssh. (Stage:
> > Connecting to Host)
> > 2011-08-30 10:21:18,834 INFO [org.nogah.bll.VdsInstaller]
> > (pool-11-thread-33) Installation of 10.x.x.x. Executing installation
> > stage. (Stage: Get the
> > unique vds id)
> > 2011-08-30 10:21:18,834 INFO
> > [org.nogah.utils.hostinstall.MinaInstallWrapper] (pool-11-thread-33)
> > Invoking /bin/echo -e `/bin/bash -c /usr/sbin/dmidecode
> > |/bin/awk ' /UUID/{ print $2; } ' | /usr/bin/tr '
> > ' '_' && cat /sys/class/net/eth*/address |/bin/sort | /usr/bin/head
> > --lines=1` on 10.x.x.x
> > 2011-08-30 10:21:18,956 INFO [org.nogah.bll.VdsInstaller]
> > (pool-11-thread-33) Installation of 10.x.x.x. Recieved message:
> > 4C4C4544-0039-3610-8057-C4C
> > 04F425131_
> > . FYI. (Stage: Get the unique vds id)
> > 2011-08-30 10:21:18,974 INFO [org.nogah.bll.VdsInstaller]
> > (pool-11-thread-33) Installation of 10.x.x.x. Assigning unique id
> > 4C4C4544-0039-3610-8057-C4C04F425131_ to Host. (Stage: Get the unique
> > vds id)
> > 2011-08-30 10:21:18,978 ERROR
> > [org.nogah.utils.hostinstall.MinaInstallWrapper] (pool-11-thread-33)
> > cat: /sys/class/net/eth*/address: No such file or directory
> > 
> > 2011-08-30 10:21:18,978 ERROR [org.nogah.bll.VdsInstaller]
> > (pool-11-thread-33) Installation of 10.x.x.x. Error: cat:
> > /sys/class/net/eth*/address: No such file or directory
> > . (Stage: Upload Installation script to Host)
> > 2011-08-30 10:21:18,978 INFO [org.nogah.bll.InstallerMessages]
> > (pool-11-thread-33) VDS message: cat: /sys/class/net/eth*/address: No
> > such file or directory
> >
Comment 16 Dan Kenigsberg 2011-08-30 21:11:15 UTC 
The code that shows in this log (/sys/class/net/eth*/address etc) is exactly the one that we dropped in http://gerrit.usersys.redhat.com/315 .

Which version of rhev-m are you using exactly? which vdsm-bootstrap.rpm is installed there?
Comment 17 Russ Builta 2011-08-31 16:34:23 UTC 
(In reply to comment #16)
> The code that shows in this log (/sys/class/net/eth*/address etc) is exactly
> the one that we dropped in http://gerrit.usersys.redhat.com/315 .
> 
> Which version of rhev-m are you using exactly? which vdsm-bootstrap.rpm is
> installed there?

-25 or rhev-m 
vdsm-bootstrap-4.9-91.1.el6.x86_64
Comment 18 Yotam Oron 2011-09-01 09:25:27 UTC 
Russ,

Could you please attach the vdsm log at /var/log/vdsm/vdsm.log ? 

Thanks.
Comment 19 Dan Kenigsberg 2011-09-01 22:57:48 UTC 
The installation logs sit under /tmp.
Comment 20 Ayal Baron 2011-09-07 17:21:38 UTC 
Russel, any news? as I told you on the email, we still need the following logs from the host:
/var/log/vdsm/vdsm.log with relevant time - there could be 100 such logs, so need to find the one with the time when the problem occured).
The older logs are xz compressed so you need to run xzgrep or xzless.

installation logs from /tmp/vds*

Thanks.
Comment 21 Russ Builta 2011-09-07 17:27:33 UTC 
Ayal, Seems as the error is not occurring anymore. After following some of the advice from rhevm_beta I cleaned up the install and started from scratch. Installed with no problems with beta #1.
Comment 22 Ayal Baron 2011-09-07 17:47:57 UTC 
(In reply to comment #21)
> Ayal, Seems as the error is not occurring anymore. After following some of the
> advice from rhevm_beta I cleaned up the install and started from scratch.
> Installed with no problems with beta #1.

Indeed it didn't make much sense to us to begin with as it looked like you were running the old version.
Ok, thanks.
Moving back to ON_QA
Comment 24 Dan Kenigsberg 2011-11-23 10:56:11 UTC 
    Technical note added. If any revisions are required, please edit the "Technical Notes" field
    accordingly. All revisions will be proofread by the Engineering Content Services team.
    
    New Contents:
Vdsm no longer assume that NICs are named eth*. This makes Vdsm compatible with the new (optional) device naming convention.
https://fedoraproject.org/wiki/Features/ConsistentNetworkDeviceNaming
Comment 25 errata-xmlrpc 2011-12-06 07:08:33 UTC 
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

http://rhn.redhat.com/errata/RHEA-2011-1782.html