SELinux is preventing abrt-action-not from write access on the directory /usr/lib64/python3.5/site-packages/problem/__pycache__.
SELinux is preventing abrt-action-not from write access on the directory /usr/lib64/python3.5/site-packages/report/__pycache__.
SELinux is preventing abrt-action-not from write access on the directory /usr/lib64/python3.5/site-packages/reportclient/__pycache__.
SELinux is preventing abrt-action-not from write access on the directory /usr/lib64/python3.5/site-packages/report/io/__pycache__.
No regressions noted. (The python cache error noted by an anonymous reporter may be an issue with a python update, if he/she has the 3.5.3 version that was in testing earlier this month, and then was un-pushed. I was seeing the same errors due to it.)
I've catched the ball before the bound somehow about Python. Even with 3.5.2-4, it turned out that it's not indeed as you wrote, except that with 225.10 modprobe will still load Nvidia only after creating own policy... 🎪
I do not know whether there was any reason to implement in policy Nvidia proprietary drivers blockings when other errors peacefully wander there from update to upgrade.
But if there were something like codenames, this issue could be called:
This update has been submitted for testing by lvrabec.
WFM
jwrdegoede edited this update.
MariaDB Cracklib plugin works well with this update.
NVIDIA proprietary driver's kernel module blocking:
An alert message during relabeling filesystem shows:
And then a sequnce of similar messages refering to files from Qt -debug packages.
NVIDIA not working
WFM
Seems, it somehow begins to resemble a vote of choosing between WFM and NWFM ((:))
+1 it also fixes BZ1411437
Pity, that it cannot fix yet Munich's city council ((:))
No regressions noted. (The python cache error noted by an anonymous reporter may be an issue with a python update, if he/she has the 3.5.3 version that was in testing earlier this month, and then was un-pushed. I was seeing the same errors due to it.)
@dhgutteridge
Thank you for clue about python. Yes, it's indeed as you wrote.
But all Nvidia proprietary driver users should be very upset after receiving this update.
lvrabec edited this update.
New build(s):
Removed build(s):
@dhgutteridge
I've catched the ball before the bound somehow about Python. Even with 3.5.2-4, it turned out that it's not indeed as you wrote, except that with 225.10
modprobe
will still load Nvidia only after creating own policy... 🎪This update has been pushed to testing.
Works for me
No regressions noted.
This fixes #1419944 for me, +1
WFM
😏 Policy RPM
selinux-policy-3.13.1-225.10.fc25.noarch
Last Seen 2017-02-23 05:06:47 EET:
Last Seen 2017-02-23 06:49:47 EET:
Last Seen 2017-02-23 23:10:27 EET:
Last Seen 2017-02-23 23:10:33 EET:
Last Seen 2017-02-23 23:10:33 EET:
Last Seen 2017-02-23 23:10:43 EET:
& my-modprobe for Nvidia ))
Looks OK here on x86_64.
works for me so far, no regressions noted
no new alerts noticed. seems to work for me.
This update has been submitted for stable by bodhi.
This update has been pushed to stable.
I do not know whether there was any reason to implement in policy Nvidia proprietary drivers blockings when other errors peacefully wander there from update to upgrade.
But if there were something like codenames, this issue could be called:
'We're so up to speed on f***ing!'
So, do you know why I'm here?
Let me to infor you: AFTER the last update my screen looks like a 950x600 on a fullhd monitor.
1) I though about an nvidia driver issue, then tried to launch the previous kernel 4.10 (instead of the last one, 4.11)
2) same results. Then in the terminal i tried "glxinfo -B" and discovered that instead of the nvidia driver I was on Gallium
3) Then I googled for a solution and found this: https://devtalk.nvidia.com/default/topic/996408/quadro-k620-on-fedora-25-unable-to-load-the-nvidia-drm-/?offset=10 At some point someone posted that a workaround is to put selinux in permissive mode
4) I tried and it works
5) From the same post I jump here just to discover that the issue with the nvidia driver was WELL KNOWN and beautifully ignored.
What was the purpose? Do you want so much a giant middle finger in your face?
"This update has been pushed to stable".... yes, from a bunch of idiots.
@valeriodean, I tell you a secret. I know it's too late but you should report a bug to fedora with SELinux Troublesooter or even better contribute a fix to fedora selinux-policy https://github.com/fedora-selinux/selinux-policy.git
And the most important that nobody gave negative karma because anonymous voices do not count.
You hear that @kuosmanen? You're an anonymous voice who doesn't count.
I missed @kuosmanen in history. Disclaimer: I am not selinux-policy maintainer