Bug 33140

Summary:	Booting hangs up on network, and network-up for a long time ( about 1 min and 1/4).
Product:	Mageia	Reporter:	w unruh <unruh>
Component:	RPM Packages	Assignee:	Mageia Bug Squad <bugsquad>
Status:	RESOLVED WORKSFORME	QA Contact:
Severity:	normal
Priority:	Normal	CC:	davidwhodgins
Version:	9
Target Milestone:	---
Hardware:	All
OS:	Linux
Whiteboard:
Source RPM:		CVE:
Status comment:
Attachments:	journalctl reports while my cmputer booot was hung up

Description w unruh 2024-04-25 05:37:21 CEST

Description of problem: On boot up ( with the bootup text showing since I removed the splashscreen) the system hangs up waiting for network.service and network-up.service. I notice that just before this Network Manager is started (is this now the default instead of the Network Center), although what comes up once Plasma comes up is NetworkApplet and network enter. Is this a battle between Netowrk Manager and Network Center?

Version-Release number of selected component (if applicable):
Up to date Mga9.

How reproducible: Always.


Steps to Reproduce:
1.Boot up
2.Wait
3.

Comment 1 Dave Hodgins 2024-04-25 06:17:52 CEST

Use one or the other, not both.
https://wiki.mageia.org/en/Switching_to_networkmanager

It is recommended to switch to networkmanager.

Closing as invalid as this is up to you as your own system admin to choose
which to use.

Status: NEW => RESOLVED
CC: (none) => davidwhodgins
Resolution: (none) => INVALID

Comment 2 Dave Hodgins 2024-04-25 06:21:43 CEST

*** Bug 33141 has been marked as a duplicate of this bug. ***

Comment 3 Dave Hodgins 2024-04-25 06:22:58 CEST

*** Bug 33142 has been marked as a duplicate of this bug. ***

Comment 4 w unruh 2024-04-25 06:36:28 CEST

Created attachment 14514 [details]
journalctl reports while my cmputer booot was hung up

These are outputs in journalctl -b while waiting to connect. What in the world are all those oceth-s# entries, or the radeth-c# enetries that Network Manager keeps trying to connect to. Where did they come from and what are they?

Aha-- there are about 300 entries in /etc/sysconfig/network-scripts of the form 
ifcfg-oceth-* and ifcfg-radeth-* all of which have contents of the form

DEVICE=oceth-s154282
BOOTPROTO=dhcp
ONBOOT=yes

All with dates either Mar 31 or Jan 14 this year
What could have caused this?
Was this some sort of attack attempt? Or more likely is it a bug in Mageia?

Comment 5 w unruh 2024-04-25 06:49:22 CEST

I did nothing to the system, it seems to have installed NetworkManager on its own.

And the huger bunch of ifcfg entries in /etc/sysconfig/network-scripts is certainly NOT a local setup problem. Something was dumping all those script, each of which was a "boot script" telling it to install at boot. I have no idea where I was on Mar 31 or Jan 14 that could have caused this.
Ie, Network Manager seems to have gone crazy, which is surely a bug.

Status: RESOLVED => REOPENED
Resolution: INVALID => (none)

Comment 6 sturmvogel 2024-04-25 11:55:43 CEST

Your truncated log shows that you have explicitely set your network interface as unmanaged.
That means, as already mentioned by Dave, administrator failure…

Comment 7 w unruh 2024-05-06 03:20:26 CEST

No, I certainly did not set the system to unmanaged. If it says unmanged then something in the system did it.

What happened was that something, certainly not me, put in about 100 ifcfg-?????
into /etc/sysconfig/network-scripts, and the machine was going throgh all of them trying to connect with each. Removal of them, and things returned to normal No idea what did it but I suspect Network Manager.

Things are working again.

Comment 8 Lewis Smith 2024-05-09 21:44:46 CEST

Without knowing at all how all those ifcfg entries in /etc/sysconfig/network-scripts got there, it is clear that they were the root of the problem, and Bill has sorted himself out by removing them.

Status: REOPENED => RESOLVED
Resolution: (none) => WORKSFORME