[Bug 1080832] New: Wicked is not acquiring an IPv4 address with DHCP since recent update
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832 Bug ID: 1080832 Summary: Wicked is not acquiring an IPv4 address with DHCP since recent update Classification: openSUSE Product: openSUSE Tumbleweed Version: Current Hardware: x86-64 OS: SUSE Other Status: NEW Severity: Normal Priority: P5 - None Component: Network Assignee: bnc-team-screening@forge.provo.novell.com Reporter: nwr10cst-oslnx@yahoo.com QA Contact: qa-bugs@suse.de Found By: --- Blocker: --- User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Firefox/52.0 Build Identifier: Computer comes up with no IPv4 address. It does have an IPv6 address. This is a recent change in Tumbleweed. This is also happening in Leap 15.0, starting with Build 127.1 (it was fine with Build 124.1). And Leap 42.3 on the same computer has no problem getting an IPv4 address. With Tumbleweed or Leap on a KVM virtual machine, everything is fine. But I am seeing this problem on two different physical computers (a Lenovo ThinkServer and a Dell laptop). My workaround is to assign static IPv4, which works fine. Also, NetworkManager seems okay in getting an IPv4 address. It is only "wicked" that has problems. Here is some output from: grep dhcp4 /var/log/messages 2018-02-13T07:39:48.088996-06:00 nwr8 wickedd-dhcp4[1505]: eth0: Request to acquire DHCPv4 lease with UUID a0ea825a-cf1c-0800-0906-000005000000 2018-02-13T07:39:48.811624-06:00 nwr8 wickedd-dhcp4[1505]: eth0: Committed DHCPv4 lease with address 192.168.1.70 (lease time 86400 sec, renew in 43200 sec, rebind in 75600 sec) 2018-02-13T07:58:58.900923-06:00 nwr8 wickedd-dhcp4[1505]: ni_dhcp4_device_close: timer active for eth0 2018-02-13T07:59:01.814380-06:00 nwr8 wickedd-dhcp4[11301]: eth0: Request to acquire DHCPv4 lease with UUID 25ef825a-e474-0b00-272c-000003000000 2018-02-13T07:59:17.190044-06:00 nwr8 wickedd-dhcp4[11301]: eth0: defer timeout 15 reached (state INIT) 2018-02-13T08:22:58.201139-06:00 nwr8 wickedd-dhcp4[1516]: eth0: Request to acquire DHCPv4 lease with UUID bef4825a-464f-0a00-6406-000005000000 2018-02-13T08:23:13.674490-06:00 nwr8 wickedd-dhcp4[1516]: eth0: defer timeout 15 reached (state INIT) 2018-02-13T08:26:21.679636-06:00 nwr8 wickedd-dhcp4[1516]: ni_dhcp4_device_close: timer active for eth0 The logs are from Leap 15.0 on the Lenovo. The first two log lines are from before I updated from 124.1 to 128.1 this morning. Reproducible: Always -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832#c2
--- Comment #2 from Neil Rickert
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832#c3
--- Comment #3 from Nirmoy Das
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832#c4
--- Comment #4 from Neil Rickert
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832#c5
--- Comment #5 from Nirmoy Das
Can you capture(tcpdump -i eth0 -n port 67 and port 68) dhcp request packets Sorry this is not helpful, can you try again with sudo tcpdump -i eth0 port 67 or port 68 -e -n -vv
Can you try wicked with /etc/wicked/local.xml: <config> <addrconf> <dhcp4> <create-cid>rfc2132</create-cid> </dhcp4> </addrconf> </config> it could be that the dhcp server doesnt like newer rfc4361 client-id -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832#c6
--- Comment #6 from Neil Rickert
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832#c7
--- Comment #7 from Nirmoy Das
Created attachment 768895 [details] tcpdump -i p1p2 -n port 67 and port 68 -e -vv
I've tried again with the tcpdump. I hope this is more useful.
I then tried adding that "local.xml" to "/etc/wicked". After restarting the network (systemctl restart network), still no IPv4 address. However, after a reboot, everything looked good. I repeated the reboot, and it still looked good.
Sorry Neil, I forgot to mention systemctl restart wickedd. The wicked daemon needs a restart to apply configs from /etc/wicked/local.xml. Wicked started using rfc4361 for client-id by default on ethernet which some dhcp server doesnt like even though its mandated by infiniband rfc4930. This a proper work around and will add it to wicked documentation. Thanks for reporting the bug. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832#c8
John Shand
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832#c9
--- Comment #9 from Neil Rickert
This a proper work around and will add it to wicked documentation.
Thanks. I'll copy that "local.xml" to other systems with the same problem. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832#c10
Nirmoy Das
i am also having a similar issue, the resolve.conf file seems to be the problem here as there are no listed nameservers after last update, if i use networkmanager the resolve file gets repaired. however, if i switch back to wicked, the resolve.conf has those nameservers removed.
Please create a new bug and add some information https://en.opensuse.org/openSUSE:Bugreport_wicked -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832#c11
--- Comment #11 from Nirmoy Das
i am also having a similar issue, the resolve.conf file seems to be the problem here as there are no listed nameservers after last update, if i use networkmanager the resolve file gets repaired. however, if i switch back to wicked, the resolve.conf has those nameservers removed.
also try to set A NETCONFIG_FORCE_REPLACE=yes in /etc/sysconfig/network/config Please upload /etc/resolve.conf as well. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832#c12
--- Comment #12 from John Shand
(In reply to John Shand from comment #8)
i am also having a similar issue, the resolve.conf file seems to be the problem here as there are no listed nameservers after last update, if i use networkmanager the resolve file gets repaired. however, if i switch back to wicked, the resolve.conf has those nameservers removed.
also try to set A NETCONFIG_FORCE_REPLACE=yes in /etc/sysconfig/network/config Please upload /etc/resolve.conf as well.
hi, i haven't needed to use this setting as yet, as the network issue seems to have resolved itself for now -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832
http://bugzilla.opensuse.org/show_bug.cgi?id=1080832#c13
Nirmoy Das
participants (1)
-
bugzilla_noreply@novell.com