Bug 1427187

Summary: Rolling back a checkpoint without any changes still reconnects all devices
Product: [Fedora] Fedora Reporter: Marius Vollmer <mvollmer>
Component: NetworkManagerAssignee: Beniamino Galvani <bgalvani>
Status: CLOSED EOL QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: low Docs Contact:
Priority: unspecified    
Version: 25CC: bgalvani, dcbw, fgiudici, lkundrak, thaller
Target Milestone: ---   
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: 2017-12-12 10:06:51 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:
Attachments:
Description Flags
[PATCH] checkpoint: reactivate/update connections only if necessary on rollback
none
[PATCH v2] checkpoint: reactivate/update connections only if necessary on rollback none

Description Marius Vollmer 2017-02-27 14:42:30 UTC
Description of problem:

When creating a checkpoint and then immediately rolling it back, all devices are reconnected although no actual changes need to be rolled back.

Cockpit rolls back a checkpoint when a configuration change has failed, and to make matters worse, Cockpit always takes checkpoints over _all_ devices, because why not.

This mostly doesn't matter, but because of the asynchronicity of everything, weird things are happening sometimes.

Version-Release number of selected component (if applicable):
NetworkManager-1.4.4-3.fc25.x86_64

How reproducible:
Always.

Steps to Reproduce:
1. gdbus call -y -d org.freedesktop.NetworkManager -o /org/freedesktop/NetworkManager -m org.freedesktop.NetworkManager.CheckpointCreate '[]' 0 0
(objectpath '/org/freedesktop/NetworkManager/Checkpoint/1',)
2. gdbus call -y -d org.freedesktop.NetworkManager -o /org/freedesktop/NetworkManager -m org.freedesktop.NetworkManager.CheckpointRollback /org/freedesktop/NetworkManager/Checkpoint/1


Actual results:
All interfaces are reconnected.

Expected results:
Nothing happens except the destruction of the checkpoint.

Additional info:
Cockpit should really make checkpoints just for the interfaces that are being changed.

Comment 1 Beniamino Galvani 2017-03-05 11:01:54 UTC
Created attachment 1260052 [details]
[PATCH] checkpoint: reactivate/update connections only if necessary on rollback

The attached patch avoids unnecessary operations (update and activate) on connections during a rollback.

Comment 2 Thomas Haller 2017-03-06 07:40:59 UTC
/* ... or if the applied connection differs */
+    if (!nm_connection_compare (dev_checkpoint->applied_connection,
+                                nm_active_connection_get_applied_connection (active),
+                                NM_SETTING_COMPARE_FLAG_EXACT)) {

you could instead compare the version-id of the active-connection.
The difference is, your check tests whether the applied-connection differs,
the version-id tests whether the active-connection was deactivated/reapplied in the meantime. That latter test might make more sense in this context(??).

_LOGT ("rollback: connection is active");

these log statements could mention the connection they are talking about... otherwise, the logfile may contain several "connection is active" lines, but it's a bit confusing. 

indention in
+                        _LOGD ("rollback: updating connection %s",
+                                nm_setting


otherwise, lgtm

Comment 3 Marius Vollmer 2017-03-06 08:44:10 UTC
Fyi, Cockpit is now much better behaved: https://github.com/cockpit-project/cockpit/pull/5986

Comment 4 Beniamino Galvani 2017-03-06 09:20:38 UTC
Created attachment 1260321 [details]
[PATCH v2] checkpoint: reactivate/update connections only if necessary on rollback

(In reply to Thomas Haller from comment #2)
> /* ... or if the applied connection differs */
> +    if (!nm_connection_compare (dev_checkpoint->applied_connection,
> +                                nm_active_connection_get_applied_connection
> (active),
> +                                NM_SETTING_COMPARE_FLAG_EXACT)) {
> 
> you could instead compare the version-id of the active-connection.
> The difference is, your check tests whether the applied-connection differs,
> the version-id tests whether the active-connection was deactivated/reapplied
> in the meantime. That latter test might make more sense in this context(??).

Both checks are fine, but comparing the version id is simpler and faster.


> _LOGT ("rollback: connection is active");
> 
> these log statements could mention the connection they are talking about...
> otherwise, the logfile may contain several "connection is active" lines, but
> it's a bit confusing. 
> 
> indention in
> +                        _LOGD ("rollback: updating connection %s",
> +                                nm_setting
> 

Fixed.

Comment 5 Thomas Haller 2017-03-06 10:56:15 UTC
lgtm now

Comment 7 Fedora End Of Life 2017-11-16 19:22:04 UTC
This message is a reminder that Fedora 25 is nearing its end of life.
Approximately 4 (four) weeks from now Fedora will stop maintaining
and issuing updates for Fedora 25. It is Fedora's policy to close all
bug reports from releases that are no longer maintained. At that time
this bug will be closed as EOL if it remains open with a Fedora  'version'
of '25'.

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, simply change the 'version'
to a later Fedora version.

Thank you for reporting this issue and we are sorry that we were not
able to fix it before Fedora 25 is end of life. If you would still like
to see this bug fixed and are able to reproduce it against a later version
of Fedora, you are encouraged  change the 'version' to a later Fedora
version prior this bug is closed as described in the policy above.

Although we aim to fix as many bugs as possible during every release's
lifetime, sometimes those efforts are overtaken by events. Often a
more recent Fedora release includes newer upstream software that fixes
bugs or makes them obsolete.

Comment 8 Fedora End Of Life 2017-12-12 10:06:51 UTC
Fedora 25 changed to end-of-life (EOL) status on 2017-12-12. Fedora 25 is
no longer maintained, which means that it will not receive any further
security or bug fix updates. As a result we are closing this bug.

If you can reproduce this bug against a currently maintained version of
Fedora please feel free to reopen this bug against that version. If you
are unable to reopen this bug, please file a new report against the
current release. If you experience problems, please add a comment to this
bug.

Thank you for reporting this bug and we are sorry it could not be fixed.