Bug 926909

Summary: after minimal DVD install, network is down
Product: [Fedora] Fedora Reporter: Andre Robatino <robatino>
Component: selinux-policyAssignee: Miroslav Grepl <mgrepl>
Status: CLOSED DUPLICATE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 19CC: dominick.grift, dwalsh, g.kaviyarasu, harald, johannbg, jonathan, lnykryn, metherid, mgrepl, mkolman, msekleta, notting, plautrba, satellitgo, sbueno, systemd-maint, vanmeeuwen+fedora, vpavlin, zbyszek
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2013-03-26 20:02:58 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
screenshot containing requested output none

Description Andre Robatino 2013-03-24 02:01:09 UTC 
Description of problem:
After a minimal install from the 19 Alpha TC1 DVD (either i386 or x86_64), the network is down. "systemctl start NetworkManager" fails. "systemctl start network" succeeds but any attempt to use the network gives the error "connect: Network is unreachable".

Please reassign to the correct Component if necessary.

Version-Release number of selected component (if applicable):
19.12 (from 19 Alpha TC1)

How reproducible:
always
Comment 1 Chris Lumens 2013-03-25 14:11:35 UTC 
Well, let's start with systemd.  They'll know how to help you debug the boot process so you can assign to the correct component.
Comment 2 Harald Hoyer 2013-03-25 15:51:41 UTC 
What is the output of?

# systemctl start NetworkManager.service
# systemctl status NetworkManager.service
# journalctl -u NetworkManager.service
# journalctl -ab _SYSTEMD_UNIT=NetworkManager.service 


You might want to save the output to a USB stick or similar.
Comment 3 Andre Robatino 2013-03-25 16:00:56 UTC 
Created attachment 716084 [details]
screenshot containing requested output

Should also add that I have not tried minimal netinstall yet, so the problem could also exist with that.
Comment 4 Andre Robatino 2013-03-26 01:59:13 UTC 
After reading https://lists.fedoraproject.org/pipermail/test/2013-March/114462.html , I confirmed that by using enforcing=0, the network comes up.
Comment 5 Andre Robatino 2013-03-26 02:21:36 UTC 
After bringing up the network, I did a yum update which pulled in selinux-policy and selinux-policy-targeted, version 3.12.1-23.fc19.noarch. (There was no systemd update.) A plain reboot after that has the network running. Reassigning to selinux-policy.
Comment 6 Andre Robatino 2013-03-26 09:22:32 UTC 
Something else is broken. I just did a text install (which pulls in updates by default, so selinux-policy-3.12.1-23.fc19.noarch is installed. But the network still doesn't come up unless I use "enforcing=0". In the previous install, I did a non-networked TC1 install, then used "enforcing=0" to apply all updates. I don't know why the network is up in one and not the other.
Comment 7 Miroslav Grepl 2013-03-26 10:27:39 UTC 
So it also does not work with permissive mode, right?
Comment 8 Andre Robatino 2013-03-26 10:46:50 UTC 
I'm fairly ignorant about SELinux, so I'd rather let someone else more knowledgeable run these tests. Sorry.
Comment 9 Daniel Walsh 2013-03-26 17:13:20 UTC 
Do you see any AVC's?

ausearch -m avc

Also is /var/run labeled correctly.

ls -lZ /var/run
Comment 10 Andre Robatino 2013-03-26 19:27:57 UTC 
It appears /var/run was incorrectly labeled as system_u:object_r:var_t:s0. Running "restorecon -v /var/run" relabeled it to system_u:object_r:var_run_t:s0 which is the same as on my F18 host. After a reboot, the F19 guest's network is up.
Comment 11 Daniel Walsh 2013-03-26 20:02:58 UTC 

*** This bug has been marked as a duplicate of bug 919374 ***