Bug 984812

Summary: Evolution uses S/MIME cert store password once and then fails
Product: [Fedora] Fedora Reporter: Bojan Smojver <bojan>
Component: evolution-data-serverAssignee: Matthew Barnes <mbarnes>
Status: CLOSED NOTABUG QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: low Docs Contact:
Priority: unspecified    
Version: 19CC: mbarnes, mcrha
Target Milestone: ---   
Target Release: ---   
Hardware: i686   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2013-07-16 22:38:00 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
Evo S/MIME config none

Description Bojan Smojver 2013-07-16 05:56:18 UTC
Description of problem:
After an upgrade to this version of evolution-data-server, S/MIME signing isn't working correctly any more. Evo asks for password for the NSS store once (to get the signing cert) and sends composed mail. However, on subsequent e-mail composition, it cannot send mail (red ribbon appears saying that message cannot be sent). I am running Evo with EWS back end. Reverting to previous e-d-s immediately fixes the problem.

Version-Release number of selected component (if applicable):
evolution-data-server-3.8.3-3.fc19

How reproducible:
Always.

Steps to Reproduce:
1. Configure S/MIME signing of outgoing messages.
2. Send mail.
3. Send mail again. Fails.

Actual results:
Fails to send subsequent mail.

Expected results:
Worked before.

Additional info:
FEDORA-2013-12643

Comment 1 Bojan Smojver 2013-07-16 06:01:15 UTC
Created attachment 774029 [details]
Evo S/MIME config

Comment 2 Milan Crha 2013-07-16 11:10:17 UTC
Thanks for a bug report. As I wrote in the update [1], this seems to work fine for me. Does the red ribbon say any detail why it cannot send a message? Maybe there's something on console, when you run evolution from it. Though if it's password related, then the thing I can think of is that the evolution-data-server provides evolution-source-registry process, and by the update the underlying binaries are changed. Even the Linux holds these changes usually fine, maybe it has some trouble with it. Could you try to restart all evolution processes
(ps ax | grep evolution) after the update, thus they get the right binaries, please? if you run gnome-shell, then it is possible the calendar factory process will be autostarted, thus stop it after the source registry process, to have it used the right DBus connection.

Comment 3 Milan Crha 2013-07-16 11:13:17 UTC
[1] https://admin.fedoraproject.org/updates/FEDORA-2013-12643

I forgot to add (beside the link), the update doesn't change much, it only adds a patch which is quite unrelated to your issue (it's for bug #981329 and bug #982737), thus it really seems like a runtime issue to me.

Comment 4 Bojan Smojver 2013-07-16 22:32:34 UTC
(In reply to Milan Crha from comment #2)
> Does the red ribbon say any detail why it cannot send a
> message?

I will get you a screenshot.

> Maybe there's something on console, when you run evolution from it.
> Though if it's password related, then the thing I can think of is that the
> evolution-data-server provides evolution-source-registry process, and by the
> update the underlying binaries are changed. Even the Linux holds these
> changes usually fine, maybe it has some trouble with it. Could you try to
> restart all evolution processes
> (ps ax | grep evolution) after the update, thus they get the right binaries,
> please? if you run gnome-shell, then it is possible the calendar factory
> process will be autostarted, thus stop it after the source registry process,
> to have it used the right DBus connection.

I always kill everything by hand before running such experiments, but I'll do it again, just to make sure.

Comment 5 Bojan Smojver 2013-07-16 22:38:00 UTC
I just tried all of this again and could not replicated. So, I must have been going mad. :-(

Comment 6 Milan Crha 2013-07-17 06:00:00 UTC
No problem, it could be some one time quirk or something.