Bug 2025617
| Summary: | Make timeout optional and remove arbitrary and undocumented maximum timeout for nm-online | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 8 | Reporter: | Nate <nlager> |
| Component: | NetworkManager | Assignee: | Fernando F. Mancera <ferferna> |
| Status: | CLOSED ERRATA | QA Contact: | Filip Pokryvka <fpokryvk> |
| Severity: | medium | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 8.3 | CC: | bgalvani, fpokryvk, lrintel, rkhan, sukulkar, systemd-maint-list, thaller, till, vbubela, wenliang |
| Target Milestone: | rc | Keywords: | Triaged |
| Target Release: | --- | Flags: | pm-rhel:
mirror+
|
| Hardware: | Unspecified | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | No Doc Update | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2022-11-08 10:07:31 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: | |||
|
Description
Nate
2021-11-22 15:34:32 UTC
Thomas, what is the reason for the 3600s upper boundary for the timeout? (In reply to Till Maas from comment #3) > Thomas, what is the reason for the 3600s upper boundary for the timeout? No reason, except that one hour as already very (very) long. There is no problem at all to increase the maximum timeout. I just don't think that is helpful. Also, "in order to determine if the network is in fact online before completing" is only partly correct (depending on interpretation what this means). NetworkManager-wait-online calls `nm-online --wait-for-startup` which technically waits until NetworkManager logs "startup complete", which is the point when all devices reached a state where there is nothing to progress (e.g. `nmcli device` shows no device activating state, among others). So, unless you also set very unusual timeouts in the NetworkManager profile (like high ipv4.required-timeout and ipv4.dhcp-timeout), NetworkManager will reach startup-complete after a few minutes -- even if the system is not connected/online. -- the exception is if you start a bridge/bond interface with DHCP, but without any slaves. In fact, that waits forever, but such a configuration isn't useful and likely a configuration error. So even if the user configures a timeout of one hour (or more), I don't think that it does what the user ultimately wants. In https://bugzilla.redhat.com/show_bug.cgi?id=1960116#c5, I was talking about replacing `NetworkManager-wait-online.service` with your own service that does the waiting suited for you. It's clear that NetworkManager-wait-online.service provides some generally useful functionality, but if that is not what you want, it is perfectly fine to replace it with your own service/script. Of course, it's also perfectly fine to discuss how to make NetworkManager-wait-online.service also cover that use-case, it just takes longer and we need to find an agreement what to do exactly and how to make it useful in general. 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 (NetworkManager 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-2022:7680 |