Bug 1875485 - deprecate /usr/lib/udev/rename_device and using ifcfg-rh files for renaming interfaces
Summary: deprecate /usr/lib/udev/rename_device and using ifcfg-rh files for renaming i...
Keywords:
Status: CLOSED WONTFIX
Alias: None
Product: Red Hat Enterprise Linux 8
Classification: Red Hat
Component: initscripts
Version: ---
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: rc
: 8.0
Assignee: Jan Macku
QA Contact: Daniel Rusek
Šárka Jana
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2020-09-03 15:20 UTC by Thomas Haller
Modified: 2022-09-03 07:27 UTC (History)
4 users (show)

Fixed In Version:
Doc Type: Deprecated Functionality
Doc Text:
.The `/usr/lib/udev/rename_device` utility has been deprecated The `udev` helper utility `/usr/lib/udev/rename_device` for renaming network interfaces has been deprecated.
Clone Of:
Environment:
Last Closed: 2022-09-03 07:27:30 UTC
Type: Bug
Target Upstream Version:


Attachments (Terms of Use)

Description Thomas Haller 2020-09-03 15:20:19 UTC
initscripts ships "/usr/lib/udev/rename_device", which parses ifcfg-rh files from /etc/sysconfig/network-scripts/ifcfg-* to rename interface. That is an alternative way for naming network interfaces to using udev rules or .link files (man systemd.link).


- network-scripts are already deprecated in rhel-8

- on Fedora 33, NetworkManager no longer writes ifcfg-rh by default ([1])

- on RHEL-9, NetworkManager also possibly will stop writing ifcfg-rh files by default

[1] https://fedoraproject.org/wiki/Changes/NetworkManager_keyfile_instead_of_ifcfg_rh


Moving away from ifcfg files will be very hard for NetworkManager, as this affects existing configuration and tools during upgrade. So while we might try to use it less and less, we will support it possibly for a long time (maybe "forever").

Still, eventually we should get rid of this file format. And we should stop suggesting this as a way to name network interfaces.


I am not requesting to (ever) drop support for it, but only to update documentation to mark this as deprecated and not recommended.



For example, the RHEL documentation ([2]) still mentions it in "2.2. How the network device renaming works", without discouraging this approach.

[2] https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/8/html/configuring_and_managing_networking/consistent-network-interface-device-naming_configuring-and-managing-networking#how-the-network-device-naming-works_consistent-network-interface-device-naming



This is also related to the discussion at bug 1851279

Comment 3 Thomas Haller 2021-06-22 11:22:36 UTC
Hi. Are there any news about this?

If we deprecated something, it should happen long before the thing actually gets removed (if ever).
It would be nice to get this at least by rhel-9-beta. Or even rhel-8.5.

Comment 4 Jan Macku 2021-06-22 12:03:07 UTC
Hi Thomas,
I pinged our documentarists, but it seems that they forgot about it. I could ping them again, but I didn't get any response last time.

I was thinking about If you could try ping your documentarists, since this is probably a part of documentation which they frequently work on.
And they could add "deprecated" notice there. But I don't know what would be faster.

Thanks
Jan

Comment 5 Thomas Haller 2021-06-22 12:39:26 UTC
(In reply to Jan Macku from comment #4)

Thanks Jan, I will reach out.


Just to confirm from initscripts' devel team:

You do agree with this request in principle, don't you?
If any concerns, please raise them. Thanks.

Comment 7 Jan Macku 2021-06-22 12:58:49 UTC
> You do agree with this request in principle, don't you?

Yes, we definitely agree on that and thank you for reporting this.

Comment 9 Thomas Haller 2022-02-14 14:59:37 UTC
hi. Could we still do this before rhel-9.0-GA?

(I am only talking about documenting the udev/rename_device mechanism as deprecated)

Comment 10 Jan Macku 2022-02-14 15:52:56 UTC
Hi Thomas, I still feel like this is a documentation issue. I'm not sure what I am supposed to do from source code perspective. I could try to ping our documentarists one more time.

Comment 11 Thomas Haller 2022-02-14 19:00:06 UTC
(In reply to Jan Macku from comment #10)
> Hi Thomas, I still feel like this is a documentation issue. I'm not sure
> what I am supposed to do from source code perspective. I could try to ping
> our documentarists one more time.

Sure. I also said that this is a documentation issue.

Comment 12 RHEL Program Management 2022-03-03 07:27:22 UTC
After evaluating this issue, there are no plans to address it further or fix it in an upcoming release.  Therefore, it is being closed.  If plans change such that this issue will be fixed in an upcoming release, then the bug can be reopened.

Comment 18 RHEL Program Management 2022-09-03 07:27:30 UTC
After evaluating this issue, there are no plans to address it further or fix it in an upcoming release.  Therefore, it is being closed.  If plans change such that this issue will be fixed in an upcoming release, then the bug can be reopened.


Note You need to log in before you can comment on or make changes to this bug.