[Bug 1206194] Distro migration name resolution failure

bugzilla_noreply at suse.com bugzilla_noreply at suse.com
Thu Dec 8 16:16:45 UTC 2022


https://bugzilla.suse.com/show_bug.cgi?id=1206194
https://bugzilla.suse.com/show_bug.cgi?id=1206194#c4

Keith Berger <kberger at suse.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |IN_PROGRESS

--- Comment #4 from Keith Berger <kberger at suse.com> ---
I confirmed the workaround by Kirk does resolve this. The root cause seems to
be during the reboot, the prior resolv.conf from the 12-sp5 which originally
has the correct entries i.e.

search iflnnm1lauhu3lf5mva5abcjsb.bx.internal.cloudapp.net
nameserver 168.63.129.16


gets reset to a "blank" file without these entries.  Once this happens and the
suse-migration-services is run, it fails to copy over the file, 

https://github.com/SUSE/suse-migration-services/blob/master/suse_migration_services/units/setup_host_network.py#L54

https://github.com/SUSE/suse-migration-services/blob/master/suse_migration_services/units/setup_host_network.py#L151

Not copying is not the root cause as the migration image still gets the correct
settings 

migration at localhost:~> ls -al /etc/resolv.conf
lrwxrwxrwx 1 root root 26 Dec  8 14:48 /etc/resolv.conf ->
/run/netconfig/resolv.conf


migration at localhost:~> cat  /run/netconfig/resolv.conf
### /etc/resolv.conf is a symlink to /var/run/netconfig/resolv.conf
### autogenerated by netconfig!
#
# Before you change this file manually, consider to define the
# static DNS configuration using the following variables in the
# /etc/sysconfig/network/config file:
#     NETCONFIG_DNS_STATIC_SEARCHLIST
#     NETCONFIG_DNS_STATIC_SERVERS
#     NETCONFIG_DNS_FORWARDER
# or disable DNS configuration updates via netconfig by setting:
#     NETCONFIG_DNS_POLICY=''
#
# See also the netconfig(8) manual page and other documentation.
#
### Call "netconfig update -f" to force adjusting of /etc/resolv.conf.
search iflnnm1lauhu3lf5mva5abcjsb.bx.internal.cloudapp.net
nameserver 168.63.129.16


HOWEVER, when we go check to see if the system is registered we run 


https://github.com/SUSE/suse-migration-services/blob/master/suse_migration_services/suse_connect.py#L38

chroot /etc/system-root SUSEConnect --list-extensions

but since the resolv.conf in /etc/system-root we get the name resolution error

Calling: ['chroot', '/system-root', 'SUSEConnect', '--list-extensions']
SUSEConnect error: SocketError: getaddrinfo: Temporary failure in name
resolution


A simple fix is instead of chroot, just use

SUSEConnect --list-extensions --root /etc/system-root 

or in terms of the code

        extensions_cmd_result = Command.run(
            ['SUSEConnect', '--list-extensions', '--root', root_path ],
            raise_on_error=False
        )


The big question is what changed that results in the resolv.conf being reset
and why are we not seeing this in the QA tests.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.suse.com/pipermail/dms-bugs/attachments/20221208/d04f3baf/attachment.htm>


More information about the Dms-bugs mailing list