Bug 2051588
| Summary: | Unavailable network after re-loading kernel modules from driverdisk | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 9 | Reporter: | Jan Stodola <jstodola> |
| Component: | dracut | Assignee: | dracut-maint-list |
| Status: | CLOSED WONTFIX | QA Contact: | qe-baseos-daemons |
| Severity: | high | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 9.0 | CC: | jikortus, jkonecny, lrintel |
| Target Milestone: | rc | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2023-08-07 07:28:19 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: | |||
| Bug Blocks: | 1942219, 2027125 | ||
|
Description
Jan Stodola
2022-02-07 14:59:08 UTC
Hi Lukas, I would like to get NM recommendation about this bug. To summarize this issue: - During boot a user will use Driver Update Disks to change the kernel driver for ethernet device (e1000 in this case) - During the load of the driver we are loosing the ethernet device and NM is disconnected - When new driver is loaded then NM will not take it back and we are not able to reach resources on the network Do you have an idea how to solve this issue? I'm thinking if I shouldn't just restart the NM service after Driver Disk load every time. Another problem is that I can't for sure tell what are the network drivers or if the driver loaded doesn't have dependency on that. And I think that restarting the NM service could load again as with the first start so we will continue as before. Could you please confirm my assumption or give me recommendation how to solve this? (In reply to Jiri Konecny from comment #2) > Hi Lukas, > > I would like to get NM recommendation about this bug. I'm looking into this. I believe this can't be solved without changes to NM, but I still need to figure out what precisely is needed. I think we need to stop the network-online.target and restart the online check again once things settle, so that the target would eventually be reached again. Unfortunatelly NM's idea of when things are settled is essentially tailored toward the initial startup; there's no way of making sure NetworkManager has seen and configured all the devices it's supposed to after we re-add the module. I'll update this bug after I'm able to crank out something useful. Uh, ok, I overthought the above a bit. Turns out the network root handling is really a best effort thing really and the whole idea is that if we try hard enough we'd eventually end up with network configured. And it appears in RHEL-9 we no longer try hard enough. This really is a dracut bug and this should solve the problem: https://github.com/dracutdevs/dracut/pull/2134 Thanks a lot Lubomir for helping us here! Switching to Dracut to get it solved correctly by the PR above. 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. |