[Sugar-devel] Bug 1752550
James Cameron
quozl at laptop.org
Mon Sep 16 17:47:33 EDT 2019
On Mon, Sep 16, 2019 at 08:49:20AM -0700, Thomas Gilliard wrote:
> FE submitted soas livef31 beta:
> does not login to liveuser
Hope I got this right.
It was difficult to understand what you meant; I didn't know what FE
was, nor livef31. I remembered that liveuser is the user name of the
automatic login of the graphical environment on a live system.
I remembered a log fragment from the #sugar-meeting channel when I was
checking back on anything after the oversight board meeting;
http://meeting.sugarlabs.org/sugar-meeting/2019-09-12#i_2968010
[00:47:45] <satellit_> walerbender MrBIOS SOAS arm and live Fedora-SoaS-Live-x86_64-31_Beta-1.1.iso do not login to liveuser
[00:48:28] <satellit_> peter robinson says to report to developers...
[00:49:02] <satellit_> no go meetin fo beta is today
[00:49:18] <satellit_> meeting for *
[00:49:19] <meeting> satellit_: Error: "for" is not a valid command.
[00:51:47] <satellit_> https://fedoraproject.org/wiki/Test_Results:Fedora_31_Beta_1.1_Installation#Default_boot_and_install_.28x86_64.29
I looked at the bug
https://bugzilla.redhat.com/show_bug.cgi?id=1752550
> Description of problem:
> soas live unable to login to liveuser
>
> Version-Release number of selected component (if applicable):
> f31 beta
Now that I know Fedora 31 beta SoaS doesn't work, I'm not going to
waste my costly data to download it. Thanks for lettting us know
about this. If the final Fedora 31 also doesn't work, we can remove
the links to it or add a warning about it.
Thinking about the problem you report, there are several mysteries you
could give more detail about; both here and in the bug report;
1. How reproducible: does it happen every time, or only sometimes?
2. Steps to Reproduce: although it may seem obvious, it is worth
writing down; e.g. "boot the image",
3. Actual results: any interesting display transitions, and something
like "does not show the desktop".
4. Expected results: something like "show the desktop".
5. Additional info: here's where the greatest mystery is; the
software tries to do something, but fails, so presumably there is
something in the systemd journal, /var/log, ~/.local or
~/.sugar/default/logs logs that explains the nature of the problem.
You or someone else will have to switch to text console and explore
the state of the system at the point of failure.
Speculating: this problem may be triggered by packaging of Sugar 0.114
without any testing. I've looked at the Fedora package sources Git,
for the sugar, sugar-toolkit-gtk3, sugar-artwork, and sugar-datastore
packages, and they moved to 0.114 about 19 days ago, and 0.114 was
released in May. Several problems were identified in 0.114 and fixed,
and the latest version is 0.116. The problems are listed in the
commits to our sugarlabs master branch for each component. If I had the
systemd journal or logs in a GitHub issue or bug report, I may
recognise the problem as one already solved.
Since 0.114 has worked for me on Debian and Ubuntu, my guess is this
problem you report is unique to Fedora.
If the problem is new as of 19 days ago, then it looks like the
packages of Sugar 0.114 are the cause, and they ought to have been
tested before being used to make an image.
--
James Cameron
http://quozl.netrek.org/
More information about the Sugar-devel
mailing list