armv7 virtual machine boot?
by Gerd Hoffmann
Hi,
Where do we stand in terms of armv7 virtual machine boot loaders?
armv8 just works very nicely, thanks to efi and grub2.
armv7 has efi firmware too, but fedora seems not not want go that route.
Last time I asked people didn't want make virtual machines boot
different (efi+grub2) than all those armv7 boards (u-boot).
So, do we have some reasonable solution to boot armv7 vms in fedora?
Without having to copy the kernel from the disk, as the fedora wiki
suggests, b/c that makes kernel updates a PITA? And preferably
something which works with virtio storage instead of slow sdcard
emulation?
What are all the koji builders doing which are (or will be?) moved to
armv7 guests on armv8 hosts?
thanks,
Gerd
7 years, 1 month
Possible Hangups on Raspberry Pi with FC25 Server
by petrenz.a@gmail.com
Hello,
I recently bought a Raspberry Pi and I want to use it as a small server. I installed FC25 on it. The only additional piece of hardware is a bluetooth keyboard/mouse device, which is connected via usb dongle. On the software side I only installed a samba server and opened the related ports with a firewall command.
This installation is working and I can access the shared folder. However after a while - it differs - the server isn't available anymore. No ping is possible, also there is no HDMI signal anymore - the only thing I can do is restart it. I'm keeping the software up to date, but that doesn't solve the problem.
So what I would like to ask is, if there are known stability issues? Or is there something I can check and maybe fix? Maybe I could provide some details (not sure what - log doesn't tell anything usefull) to track down any issue.
Best Regards
Alex
7 years, 1 month
Fedora Server image - where is newaliases
by Robert Moskowitz
I am building a Fedora Server image. As such, I am modifying the
/etc/aliases file and need to run newaliases. I am use to that always
being part of a Centos base image, it is not in the Fedora Server base
image, it seems.
I tried 'dnf whatprovides newaliases' and got 'No Matches Found'.
But then my search foo is notoriously weak.
Can someone please point me to the proper rpm to install so that I it?
thanks
7 years, 1 month
Fedora-Server-armhfp-25-1.3-sda - httpd userdir and selinux probem
by Robert Moskowitz
I did all of my steps to build a Fedora Server with updates and
installed httpd.
I edited the /etc/httpd/conf.d/userdir.conf to enable userdir. I kept
all of the rest of this conf file unchanged.
I added my z00-init.conf file:
ServerAdmin rgm(a)htt-consult.com
ServerName medon.htt-consult.com:80
# NameVirtualHost *:80
# NameVirtualHost *:443
I followed the permission instructions in the userdir.conf file, copied
some files into my public_html
set my IP addr and all the rest, put the Cubie on my DMZ and tried to
access a file:
http://medon.htt-consult.com/~rgm/cubieboard/cubietower-2.JPG
This works. You can try this also, as the server is public.
But I cannot access just a directory:
http://medon.htt-consult.com/~rgm/cubieboard/
I get:
=====
Forbidden
You don't have permission to access /~rgm/cubieboard/ on this server.
=====
If I disable SELinux with 'setenforce 0' I get the listing of files in
this directory.
The userdir.conf default is:
<Directory "/home/*/public_html">
AllowOverride FileInfo AuthConfig Limit Indexes
Options MultiViews Indexes SymLinksIfOwnerMatch IncludesNoExec
Require method GET POST OPTIONS
</Directory>
So this should work. /var/log/httpd/error.log reports:
[Sat Feb 04 23:24:13.289963 2017] [negotiation:error] [pid 865]
(13)Permission denied: [client 192.168.160.12:56918] AH00686: cannot
read directory for multi: /home/rgm/public_html/cubieboard/
permissions are:
# ls -ls /home/rgm/public_html/
total 4
0 drwxrwxr-x. 2 rgm rgm 54 Feb 4 22:51 cubieboard
0 drwxrwxr-x. 2 rgm rgm 203 Feb 4 22:51 Harpsichord
0 drwxrwxr-x. 2 rgm rgm 73 Feb 4 22:51 ietf
4 -rw-rw-r--. 1 rgm rgm 12 Jan 21 2013 index.html
# ls -ls /home/rgm/public_html/cubieboard/
total 3504
1420 -rw-rw-r--. 1 rgm rgm 1450256 Aug 27 2014 cubietower-2.JPG
2084 -rw-r--r--. 1 rgm rgm 2131328 Sep 4 2014 cubietower-3.JPG
Why did I suspect SELinux you ask? Well I get all these warnings when I
do any SELinux command. What follows is everything I could gather from
the console.
So please tell me what is the problem and how to fix it. I really want
to get this simple server up and in production already...
Upgrading : selinux-policy-targeted-3.13.1-225.6.fc25.noarch
310/844
[ 8981.136506] SELinux: Permission validate_trans in class security not
defined in policy.
[ 8981.144734] SELinux: Permission module_load in class system not
defined in policy.
[ 8981.152998] SELinux: the above unknown classes and permissions will
be allowed
[ 8985.835670] SELinux: Context
unconfined_u:unconfined_r:sandbox_t:s0-s0:c0.c1023 became invalid
(unmapped).
[ 8987.072186] SELinux: Context
system_u:unconfined_r:sandbox_t:s0-s0:c0.c1023 became invalid (unmapped).
/var/lib/selinux/targeted/active/policy.kern: read error
(tried to read 8192 bytes from offset 3848760)
cannot reconstruct rpm from disk files
[ 0.006234] SELinux: Initializing.
[ 6.113236] systemd[1]: systemd 231 running in system mode. (+PAM
+AUDIT +SELINUX +IMA -APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP
+GCRYPT +GNUTLS +ACL +XZ +LZ4 +SECCOMP +BLKID +ELFUTILS +KMOD +IDN)
[ 17.446033] SELinux: Permission validate_trans in class security not
defined in policy.
[ 17.454298] SELinux: Permission module_load in class system not
defined in policy.
[ 17.462557] SELinux: the above unknown classes and permissions will
be allowed
[ 17.544762] systemd[1]: Successfully loaded SELinux policy in 910.703ms.
Upgrading : container-selinux-2:2.5-1.fc25.noarch
179/500
[11840.708423] SELinux: Permission validate_trans in class security not
defined in policy.
[11840.716674] SELinux: Permission module_load in class system not
defined in policy.
[11840.725023] SELinux: the above unknown classes and permissions will
be allowed
[11841.999564] SELinux: Context system_u:system_r:gear_t:s0-s0:c0.c1023
became invalid (unmapped).
[11847.589662] SELinux: Context
unconfined_u:system_r:gear_t:s0-s0:c0.c1023 became invalid (unmapped).
libsemanage.semanage_direct_remove_key: Removing last container module
(no other container module exists at another priority).
# semanage port -a -t ssh_port_t -p tcp 1234 /* not the real port number
[116454.082753] SELinux: Permission validate_trans in class security
not defined in policy.
[116454.091093] SELinux: Permission module_load in class system not
defined in policy.
[116454.099465] SELinux: the above unknown classes and permissions will
be allowed
# setsebool -P httpd_enable_homedirs on
[120049.339391] SELinux: Permission validate_trans in class security
not defined in policy.
[120049.347700] SELinux: Permission module_load in class system not
defined in policy.
[120049.356072] SELinux: the above unknown classes and permissions will
be allowed
thanks
7 years, 1 month
Fedora-Server-armhfp-25-1.3-sda - partition resize problem
by Robert Moskowitz
When I created the server image, I used the --resizefs option.
This seemed to work. But I put the mSD card into my notebook to copy
all of my files for the web server, and I am seeing only:
# df -h
Filesystem Size Used Avail Use% Mounted on
....
/dev/sdc4 2.8G 1.8G 1.1G 63% /run/media/rgm/__
but parted says:
# parted /dev/sdc print
Model: Generic- SD/MMC (scsi)
Disk /dev/sdc: 15.5GB
Sector size (logical/physical): 512B/512B
Partition Table: msdos
Disk Flags:
Number Start End Size Type File system Flags
1 1049kB 31.5MB 30.4MB primary fat16 lba
2 31.5MB 543MB 512MB primary ext4 boot
3 543MB 1055MB 512MB primary linux-swap(v1)
4 1055MB 15.5GB 14.5GB primary xfs
gparted information on the partition reports:
10.69 GiB of unallocated space within the partition.
To grow the file system to fill the partition, select the partition and
choose the menu item:
Partition --> Check.
For now all I am going to do is copy a few files until I get direction
here on how to fix this.
7 years, 1 month
Fedora-Xfce-armhfp-Rawhide-20170119.n.0-sda.raw.xz problems on CubieTruck
by Robert Moskowitz
I build a CubieTruck image using the installer script with the following
options:
--target=Cubietruck --media=/dev/sdb --norootpass --selinux=OFF
--addconsole --resizefs
I booted up the image with a UDB keyboard and VGA monitor attached plus
the TTY console.
After I logged into the console as root, I set the root password, and
added myself as a user. I then used this userID to log into the Xfce login.
Next I did 'dnf update' Of course this took time to run. But a VERY
long time between the cleanup and verify phases. During these I got the
message:
Image Name: 4.10.0-0.rc5.git0.1.fc26.armv7hl
Created: Wed Jan 25 10:04:59 2017
Image Type: ARM Linux Kernel Image (uncompressed)
Data Size: 6054568 Bytes = 5912.66 kB = 5.77 MB
Load Address: 00008000
Entry Point: 00008000
Image Name: initramfs
Created: Wed Jan 25 10:05:01 2017
Image Type: ARM Linux RAMDisk Image (uncompressed)
Data Size: 16525173 Bytes = 16137.86 kB = 15.76 MB
Load Address: 00000000
Entry Point: 00000000
dracut: Disabling early microcode, because kernel does not support it.
CONFIG_MICROCODE_[AMD|INTEL]!=y
eventually the update finished. The system SEEMS to be working, but
what is with this message?
thank you
7 years, 1 month