Does anyone know why pam_pkcs11 is missing in the Fedora 25 repos?
Systems upgraded from 24 still have the 24 version and it appears to
work (at least as well as it did in 24). Newly install 25 systems
don't appear to be able to install it.
I don't see anything in the release notes or common bugs about this
and /etc/pam.d/smartcard-auth still refers to it.
I burn the iso or 25 to usb and I also did it to a dvd same error . When I
click the install I get the the part to select English . Then I click
continue . Then the install locks up it will not get past the point at all
it just set there . I also have try to re down it same thing i have tried
more then one usb drive And dvd disk
How can I tell my desktop (fedora-23/xfce4) to use Emacs to open
any file with a ".org" extension? Everything I've found on the
internet seems to set an application based on mimetype but ".org"
files are seen as "text/plain" (at least by the 'file' program)
and I don't want to open all text files with Emacs, just ".org"
In it's infinite wisdom, the upgrade from F23 to F24 updates PostgreSQL from
9.4 to 9.5.5 making existing databases unusable.
I suppose that there is a warning about this somewhere but it certainly isn't
part of the upgrade process.
I know that I can use pg_upgrade to migrate my databases but need the 9.4
binaries to do that.
Are the files in /usr/lib64/pgsql/postgresql-9.4/bin the correct ones to use
This was weird. I hit Ctrl-Alt-F2 to get a vterm, got a login prompt,
logged in as root, was about to type something when the screen
goes absolutely black. Ctrl-Alt-F3 - still black, Ctrl-Alt-F1,
It was like using a vterm crashed the nvidia driver or something
(I am using the rpmfusion nvidia driver, not nouveau because
nouveau kept freezing up on me).
Anyone else ever seen this happen?
I was able to ssh in from my phone and reboot there :-).
I'm testing VMware Workstation using a Windows 10 guest, but having
problems with mouse integration. Within the guest the mouse cursor
moves and I can click once, to open a settings option for example, but
subsequent clicks are ignored until I force close the window with Alt-
F4, whereupon I can do it again. I've tried both my normal Bluetooth
mouse and a Microsoft wireless mouse (both of which work correctly in
Fedora) but with the same result. My wireless keyboard works correctly
in both host and guest. Switching the mouse between host and guest also
The VMware documentation has various suggestions but mainly they seem
to involve mounting the USB filesystem (usbfs) under /proc/bus/usb,
which doesn't exist in Fedora.
I'm open to ideas on either the specific mouse issue or on the wider
Trying to get input regarding nfs (erver/client) and stale file handles..
Research indicates the stale file issue can arise when the nfs file
changes and the client attempts to interact with an older/out of date
The usual solutions have information on doing a remount process on
either the client or serverside of the nfs..
Is this what you guys have seen?
In my use case as I have a bunch of clients interacting/modifying with
the nfs files might be difficult to refresh on the serverside due to
conflicts from the clients using the nfs files..
On the clientside easy enough to automate a process/logic to every so
often do a umount/remount process on the individual client.
Two headline features of Fedora 25 are:
- Wayland by default
- Initial support for PRIME offload rendering for NVIDIA Optimus
My laptop has a dual GPU setup (Skylake + GTX 970), but I can't seem to
start a Wayland session at all. GDM even runs using Xorg, and the
"GNOME" session logs in using Xorg.
I checked my initramfs, but it includes the drm module, as well as both
kernel modules: nouveau.ko and i915.ko. From a console session, I can
start weston (weston-launch), but I can't start gnome-shell (dbus-run-
session -- gnome-shell --display-server --wayland). It errors with:
mutter-WARNING **: Can't initialize KMS backend: could not find drm kms
The Wayland session worked in F24, so I'm guessing this is because of
the dual-GPU support enabled in F25. Should these two features be
mutually exclusive? Should I expect wayland *or* GPU render offloading?
Maybe too much food over the weekend.. Maybe brain cells are fried!
Created a basic/simple 5 line php test, It runs so chown/chmod are ok.
However, when I then run it as a background process, and check the
procTBL, the PID is there.. the process never leaves, unless I
manually kill it.. so... what have I forgotten.. I know.. the homer
simpson (doh!!) is going to be painful on this one..
my test php
print "running child_test \n";
running from the cmd line..
/cloud_nfs/child_test.php << runs fine, and not in procTBL after running..
trying to background from the cmd line..
/cloud_nfs/child_test.php & << never goes from procTBL unless
/cloud_nfs/child_test.php 2>&1 & << never goes from procTBL unless
So.. what did I screw up to run in background, and be removed from procTBL..