Bug 1817236
| Summary: | non_virtual_ip fails with unexpected keyword argument | ||
|---|---|---|---|
| Product: | OpenShift Container Platform | Reporter: | Ben Nemec <bnemec> |
| Component: | Machine Config Operator | Assignee: | Ben Nemec <bnemec> |
| Status: | CLOSED ERRATA | QA Contact: | Victor Voronkov <vvoronko> |
| Severity: | high | Docs Contact: | |
| Priority: | high | ||
| Version: | 4.4 | CC: | smilner, vvoronko |
| Target Milestone: | --- | ||
| Target Release: | 4.5.0 | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: |
Cause: Some routes could not be handled by the non_virtual_ip parsing code.
Consequence: Services that need to be configured with the non-virtual ip will fail.
Fix: Make non_virtual_ip parsing code more flexible.
Result: Routes are parsed and services are configured correctly.
|
Story Points: | --- |
| Clone Of: | Environment: | ||
| Last Closed: | 2020-07-13 17:23:43 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: | |||
After fixing the expires problem, I'm seeing another failure on the following route: default proto ra metric 100 \ nexthop via fe80::69e0:8df7:4351:4eb6 dev enp1s0 weight 1 \ nexthop via fe80::f620:b5b3:612a:8446 dev enp1s0 weight 1 \ nexthop via fe80::b449:f968:42ce:eef5 dev enp1s0 weight 1 pref medium This one is because of the \'s in the route, which are standing in for line breaks. The nexthop stuff isn't handled correctly either, so I think we might want to allow this class to take arbitrary params and just ignore the ones we don't care about. The fix for this ended up going in as part of https://github.com/openshift/machine-config-operator/pull/1601 *** Bug 1821950 has been marked as a duplicate of this bug. *** Verified on 4.5.0-0.nightly-2020-04-14-031010 by adding multiple lines route to the master: sudo ip -6 r add f000::/64 proto ra metric 100 nexthop via fe80::1 dev enp5s0 nexthop via fe80::2 dev enp5s0 # ip -o -6 r f000::/64 proto ra metric 100 \ nexthop via fe80::1 dev enp5s0 weight 1 \ nexthop via fe80::2 dev enp5s0 weight 1 pref medium ... [core@master-0-0 ~]$ non_virtual_ip fd2e:6f44:5dd8:c956::2 Filtering out Address(127.0.0.1/8, dev=lo) due to it having host scope Filtering out Address(::1/128, dev=lo) due to it having host scope Filtering out Address(fd2e:6f44:5dd8::5/128, dev=enp5s0) due to it being deprecated Filtering out Address(fd2e:6f44:5dd8::2/128, dev=enp5s0) due to it being deprecated Checking V6Route(f000::/64, dev=enp5s0) for Address(fd2e:6f44:5dd8::11a/128, dev=enp5s0) Checking V6Route(fd2e:6f44:5dd8::/64, dev=enp5s0) for Address(fd2e:6f44:5dd8::11a/128, dev=enp5s0) Is fd2e:6f44:5dd8:c956::2 between fe80:: and fe80::ffff:ffff:ffff:ffff Is fd2e:6f44:5dd8:c956::2 between fd2e:6f44:5dd8:: and fd2e:6f44:5dd8:0:ffff:ffff:ffff:ffff Is fd2e:6f44:5dd8:c956::2 between fd2e:6f44:5dd8::11a and fd2e:6f44:5dd8::11a Is fd2e:6f44:5dd8:c956::2 between fe80:: and fe80::ffff:ffff:ffff:ffff Is fd2e:6f44:5dd8:c956::2 between fe80:: and fe80::ffff:ffff:ffff:ffff Is fd2e:6f44:5dd8:c956::2 between fe80:: and fe80::ffff:ffff:ffff:ffff Is fd2e:6f44:5dd8:c956::2 between fd99:: and fd99::ffff:ffff:ffff:ffff Is fd2e:6f44:5dd8:c956::2 between fd01:0:0:2:: and fd01::2:ffff:ffff:ffff:ffff Is fd2e:6f44:5dd8:c956::2 between fe80:: and fe80::ffff:ffff:ffff:ffff [core@master-0-0 ~]$ cat /etc/resolv.conf # Generated by KNI resolv prepender NM dispatcher script search ocp-edge-cluster-0.qe.lab.redhat.com nameserver fd2e:6f44:5dd8:0:0:0:0:2 nameserver fe80::5054:ff:fe49:76e3%enp5s0 nameserver fd2e:6f44:5dd8::1 sudo journalctl -b | grep prepender ... Apr 20 09:12:06 master-0-0.ocp-edge-cluster-0.qe.lab.redhat.com root[3861282]: NM resolv-prepender triggered by enp4s0 dhcp6-change. Apr 20 09:12:06 master-0-0.ocp-edge-cluster-0.qe.lab.redhat.com nm-dispatcher[3861275]: <13>Apr 20 09:12:06 root: NM resolv-prepender triggered by enp4s0 dhcp6-change. Apr 20 09:12:06 master-0-0.ocp-edge-cluster-0.qe.lab.redhat.com root[3861283]: NM resolv-prepender: Prepending 'nameserver fd2e:6f44:5dd8:0:0:0:0:2' to /etc/resolv.conf (other nameservers from /var/run/NetworkManager/resolv.conf) Apr 20 09:12:06 master-0-0.ocp-edge-cluster-0.qe.lab.redhat.com nm-dispatcher[3861275]: <13>Apr 20 09:12:06 root: NM resolv-prepender: Prepending 'nameserver fd2e:6f44:5dd8:0:0:0:0:2' to /etc/resolv.conf (other nameservers from /var/run/NetworkManager/resolv.conf) 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. https://access.redhat.com/errata/RHBA-2020:2409 |
Description of problem: In an ipv6 deployment, non_virtual_ip may fail with the following stack trace: Traceback (most recent call last): File "/usr/local/bin/non_virtual_ip", line 235, in <module> main() File "/usr/local/bin/non_virtual_ip", line 220, in main iface_addrs = list(interface_addrs(filters)) File "/usr/local/bin/non_virtual_ip", line 163, in interface_addrs for route in (V6Route.from_line(rline) for rline in route_out.splitlines()): File "/usr/local/bin/non_virtual_ip", line 163, in <genexpr> for route in (V6Route.from_line(rline) for rline in route_out.splitlines()): File "/usr/local/bin/non_virtual_ip", line 81, in from_line return cls(**attrs) TypeError: __init__() got an unexpected keyword argument 'expires' This is because there is a route that looks like this: fd00:1101::/64 dev enp1s0 proto kernel metric 256 expires 5sec pref medium non_virtual_ip does not know how to handle the expires field so it fails when it gets to this route. I'm unsure how common this would be in regular deployments because I'm working on some semi-related changes that may be triggering this more often, but since that is a valid route we should make sure that the script can handle it.