RHEL Engineering is moving the tracking of its product development work on RHEL 6 through RHEL 9 to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "RHEL project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs in the statuses "NEW", "ASSIGNED", and "POST" are being migrated throughout September 2023. Bugs of Red Hat partners with an assigned Engineering Partner Manager (EPM) are migrated in late September as per pre-agreed dates. Bugs against components "kernel", "kernel-rt", and "kpatch" are only migrated if still in "NEW" or "ASSIGNED". If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "RHEL project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/RHEL-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.
Bug 1773637 - cloud-init may rewrite network interface config during subsequent reboots if config-drive is not detected.
Summary: cloud-init may rewrite network interface config during subsequent reboots if ...
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Red Hat Enterprise Linux 7
Classification: Red Hat
Component: cloud-init
Version: 7.0
Hardware: x86_64
OS: Linux
high
high
Target Milestone: rc
: ---
Assignee: Eduardo Otubo
QA Contact: xiachen
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2019-11-18 15:27 UTC by Matt Flusche
Modified: 2024-11-20 07:54 UTC (History)
12 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2020-11-05 12:20:01 UTC
Target Upstream Version:
Embargoed:

Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Knowledge Base (Solution) 4871261 0 None None None 2020-03-03 21:31:13 UTC

Description Matt Flusche 2019-11-18 15:27:40 UTC 
Description of problem:

For unknown reasons during boot, cloud-init may fail to detect a config-drive and reset the networking to the initial (first-boot) config.

Reference for this issue:  https://bugzilla.redhat.com/show_bug.cgi?id=1760806

It would seem that cloud-init should still detect it is not a first boot based on the contents of /var/lib/cloud/instance and skip the network config.


Version-Release number of selected component (if applicable):
cloud-init-18.5-3.el7.x86_64

How reproducible:
100%

Steps to Reproduce:
1. create cloud instance with config-drive.
2. modify config of primary interface.
3. reboot and verify config of interface does not change; should see messages like the following in /var/log/cloud-init.log:

2019-11-18 14:45:47,532 - stages.py[DEBUG]: No network config applied. Neither a new instance nor datasource network update on 'System boot' event

4. remove cloud-init partition.
5. reboot and verify cloud-init rewrites interface's networking config.

Actual results:
contents of /etc/sysconfig/network-scripts/ifcfg-XX are modified to the default, first-boot, config.

Expected results:
No change to /etc/sysconfig/network-scripts/ifcfg-XX  after first boot.

Additional info:
Comment 2 David Hill 2020-03-03 21:15:36 UTC 
The problem here is that if the datasource is not available, we fallback to Fallback/None which changes the instance-id to iid-datasource-none [1]:

~~~

   def is_new_instance(self):
        previous = self.previous_iid()
        ret = (previous == NO_PREVIOUS_INSTANCE_ID or
               previous != self.datasource.get_instance_id())
        return ret
~~~
and previous_iid is defined by this:
~~~

    def previous_iid(self):
        if self._previous_iid is not None:
            return self._previous_iid

        dp = self.paths.get_cpath('data')
        iid_fn = os.path.join(dp, 'instance-id')
        try:
            self._previous_iid = util.load_file(iid_fn).strip()
        except Exception:
            self._previous_iid = NO_PREVIOUS_INSTANCE_ID

        LOG.debug("previous iid found to be %s", self._previous_iid)
        return self._previous_iid
~~~

This is super easy to reproduce, attach a CD containing the metadata, boot a new instance.  Once it booted, shut it down, detach the CD and reboot it again.   This is what's happening on an atomic host [2] and this as for effect of wiping the network configuration as well as reverting many settings to the default settings.


[1] https://cloudinit.readthedocs.io/en/latest/topics/datasources/fallback.html
[2] https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux_atomic_host/7/html/installation_and_configuration_guide/types_of_installation#vmware_installation
Comment 5 Eduardo Otubo 2020-09-17 12:16:23 UTC 
You're using cloud-init-18.5-3.el7.x86_64, and this same bug should be fixed on cloud-init-18.5-5.el7.x86_64, please give it a try.
Comment 17 Brandy Northrop 2024-05-03 13:07:15 UTC Comment hidden (spam)
Comment 18 merry678 2024-11-09 09:51:27 UTC 
Did you have a fix on this issue?  https://www.spotify-stats.com
Comment 19 carolgrey98 2024-11-19 05:09:35 UTC 
Cloud-init fails to detect the presence of a config-drive during boot and resets the network configuration to its default first-boot state, even if the system is not in a first-boot state. This happens when the cloud-init partition is removed, causing the network settings in /etc/sysconfig/network-scripts/ifcfg-XX to be overwritten. The issue is reproducible and contradicts the expected behavior of cloud-init skipping network reconfiguration after the first boot https://www.wellstar-mychart.com
Note You need to log in before you can comment on or make changes to this bug.