Bug 1789246
| Summary: | RFE: Support NetworkManager unmanaged interface | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 8 | Reporter: | Edward Haas <edwardh> |
| Component: | nmstate | Assignee: | Gris Ge <fge> |
| Status: | CLOSED ERRATA | QA Contact: | Mingyu Shi <mshi> |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | 8.2 | CC: | danken, ferferna, fge, jiji, jishi, mburman, network-qe, phoracek, rgarcia, till |
| Target Milestone: | rc | Keywords: | FutureFeature, Triaged |
| Target Release: | 8.4 | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | nmstate-0.4.1-1.el8 | Doc Type: | If docs needed, set a value |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2021-05-18 15:16:54 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: | |||
| Bug Depends On: | 1848817 | ||
| Bug Blocks: | 1825061, 1875967, 1894575 | ||
I don't know what nispor is it why is it needed. I know that we need this functionality for testing, but I'm not aware of an urgent business requirement. Eddy may know better. I hope to see it on upstream earlier than 8.4, perhaps as a plugin in an experimental state. As we expect NM to manage all networking on the host, I think it is not an urgent production requirement (or at least I cannot think of one). Said that, it is still an issue as the expectations from the report is to "see" all the host state and not be limited by technicalities like this one. One may set an interface as not manageable by NM for debugging, and the report will wrongly show its state. Background story. The nispor is introduced in order to provide API for full network state via kernel netlink or plugin. nispor will do the parsing on network state and nmstate will use it for state query. Thanks for the info. I will move this RFE to 8.4. Gris, now with nispor in place, is there any update regarding this BZ? (In reply to Petr Horáček from comment #7) > Gris, now with nispor in place, is there any update regarding this BZ? I am still waiting RCM team to include nispor into next RHEL release where nmstate could depend on. For the recent upstream 0.4.0 release, bridge/dummy/veth/vlan/vxlan has been supported. the only remaining work are bond and SR-IOV. Thanks Gris. It has been a month, have we heard from RCM? (In reply to Petr Horáček from comment #11) > Thanks Gris. It has been a month, have we heard from RCM? Should be all set now. It works fine on regular linux network devices. veth, dummy, eth, vlan, bridge, vrf and bond were tested. Now unmanaged team and ovs are not supported, not all properties, at least. Verified with versions: NetworkManager-1.28.0-0.1.el8.x86_64 nispor-0.6.1-2.el8.x86_64 nmstate-0.4.1-2.el8.noarch DISTRO=RHEL-8.4.0-20201103.d.0 Linux hp-dl380g10-02.rhts.eng.pek2.redhat.com 4.18.0-241.el8.dt1.x86_64 #1 SMP Mon Nov 2 08:24:31 EST 2020 x86_64 x86_64 x86_64 GNU/Linux The test case added. (In reply to Mingyu Shi from comment #16) > It works fine on regular linux network devices. veth, dummy, eth, vlan, > bridge, vrf and bond were tested. > Now unmanaged team and ovs are not supported, not all properties, at least. Gris, is it intended/documented that team and ovs are not supported to be unmanaged? (In reply to Till Maas from comment #17) > (In reply to Mingyu Shi from comment #16) > > It works fine on regular linux network devices. veth, dummy, eth, vlan, > > bridge, vrf and bond were tested. > > Now unmanaged team and ovs are not supported, not all properties, at least. > > Gris, is it intended/documented that team and ovs are not supported to be > unmanaged? Yes. Supporting NetworkManager unmanaged OVS/team interface require a lot effort. Without real use case, I would like to postpone it. The document is at: * https://www.nmstate.io/devel/api.html#limitations-of-team-interface * https://www.nmstate.io/devel/api.html#limitations-of-openvswitch-interfaces and 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 (nmstate bug fix and enhancement update), 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-2021:1748 |
Description of problem: Reporting the state of an interface that is not managed by NM will not show the current IP addresses stack information, even if there is one. ``` 22: dummy99: <BROADCAST,NOARP,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN group default qlen 1000 link/ether c2:8e:cc:44:4b:20 brd ff:ff:ff:ff:ff:ff inet 1.1.1.1/24 scope global dummy99 valid_lft forever preferred_lft forever inet6 fe80::c08e:ccff:fe44:4b20/64 scope link valid_lft forever preferred_lft forever interfaces: - name: dummy99 type: dummy state: down ipv4: enabled: false ipv6: enabled: false mac-address: C2:8E:CC:44:4B:20 mtu: 1500 ``` Version-Release number of selected component (if applicable): How reproducible: 100% Steps to Reproduce: 1. Define a new dummy device using ip-link. 2. Define an IP on the dummy device using ip-address. 3. Set NM to unmanage the device using nmcli-device. 4. Report the current state of the device using nmstatectl-show Actual results: IP stack information shows as `disabled`. Expected results: IP stack information should show the actual state on the kernel. Additional info: