Bug 808203

Summary: NM only accepts first RDNSS from router advertisements and disconnects because of RDNSS timeout
Product: [Fedora] Fedora Reporter: Pavel Šimerda (pavlix) <psimerda>
Component: NetworkManagerAssignee: Dan Williams <dcbw>
Status: CLOSED DUPLICATE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: high Docs Contact:
Priority: unspecified    
Version: 17CC: danw, dcbw, deejay1
Target Milestone: ---Keywords: Reopened
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2012-04-19 21:12:51 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Pavel Šimerda (pavlix) 2012-03-29 19:38:17 UTC 
After update to NetworkManager-0.9.4.0-1.git20120328.fc17.x86_64, NM drops connection on every RDNSS timeout. Logs that previously contained records about
refreshed RDNSS records disappeared. This regression can possibly make NM unusable on networks that use RA with RDNSS records for IPv6 configuration.

# cat /etc/NetworkManager/NetworkManager.conf 
[main]
plugins=ifcfg-rh
[logging]
level=DEBUG
domains=IP4,IP6,DHCP4,DHCP6,WIFI,HW,BT,SUPPLICANT,SUSPEND,CORE,DEVICE

# grep debug /etc/rsyslog.conf 
*.*						/tmp/debug

# grep -i rdnss /tmp/debug
Mar 29 21:18:56 dragon NetworkManager[6160]: <debug> [1333048736.186648] [nm-ip6-manager.c:277] rdnss_expired(): (wlan0): IPv6 RDNSS information expired
Mar 29 21:18:56 dragon NetworkManager[6160]: <debug> [1333048736.186712] [nm-ip6-manager.c:312] set_rdnss_timeout(): (wlan0): removing expired RA-provided nameserver fe80::20c:42ff:fe13:857b
Mar 29 21:19:07 dragon NetworkManager[6160]: <debug> [1333048747.133161] [nm-ip6-manager.c:748] process_nduseropt_rdnss(): (wlan0): found RA-provided nameserver fe80::20c:42ff:fe13:857b (expires in 610 seconds)
Mar 29 21:29:17 dragon NetworkManager[6160]: <debug> [1333049357.186226] [nm-ip6-manager.c:277] rdnss_expired(): (wlan0): IPv6 RDNSS information expired
Mar 29 21:29:17 dragon NetworkManager[6160]: <debug> [1333049357.186317] [nm-ip6-manager.c:312] set_rdnss_timeout(): (wlan0): removing expired RA-provided nameserver fe80::20c:42ff:fe13:857b
Mar 29 21:29:28 dragon NetworkManager[6160]: <debug> [1333049368.554401] [nm-ip6-manager.c:748] process_nduseropt_rdnss(): (wlan0): found RA-provided nameserver fe80::20c:42ff:fe13:857b (expires in 610 seconds)

Related to bug 753482 but much worse.
Comment 1 Pavel Šimerda (pavlix) 2012-03-30 05:56:34 UTC 
I forgot to add that this was experienced only on wireless so far.
Comment 2 Łukasz Jernaś 2012-04-03 16:19:12 UTC 
Happens also on F16 even with wired connetions with IPv6 enabled
Comment 3 Pavel Šimerda (pavlix) 2012-04-03 16:29:50 UTC 
Hmm, now I actually have some positive messages:

Apr  3 08:22:39 dragon NetworkManager[697]: <debug> [1333434159.114071] [nm-ip6-manager.c:730] process_nduseropt_rdnss(): (eth0): refreshing RA-provided nameserver fe80::20c:42ff:fe13:857b (expires in 1810 seconds)

Still NetworkManager-0.9.4.0-1.git20120328.fc17.x86_64, kernel-3.3.0-1.fc17.x86_64

I don't know what makes the difference.
Comment 4 Dan Winship 2012-04-19 14:37:42 UTC 

*** This bug has been marked as a duplicate of bug 785772 ***
Comment 5 Pavel Šimerda (pavlix) 2012-04-19 20:39:46 UTC 
I believe this bugreport was closed duplicate by mistake. I don't see any
relation between this bug and bug 785772.
Comment 6 Dan Winship 2012-04-19 21:12:51 UTC 
Yeah, sorry, I should have explained. NM's behavior in 785772 causes the kernel to discard every RA after the first one, so NM never sees another RDNSS notification after that and so is guaranteed to hit the timeout.

*** This bug has been marked as a duplicate of bug 785772 ***
Comment 7 Pavel Šimerda (pavlix) 2012-04-19 21:14:23 UTC 
Ok then, but please don't close the other one.