The officially official Devuan Forum!

You are not logged in.

#226 Other Issues » Beowulf - Wordpress and PHP 7.3 » 2019-03-25 16:20:45

Geoff 42
Replies: 1

I had been having a quick look at WordPress, to understand what my colleagues were talking about!
I had got it running under ASCII, but after upgrading to Beowulf, it stopped working, just displaying PHP code.
I believe that the problem was caused by going from PHP 5 to PHP 7.3. This was fixed by enabling the php7.3 module

a2enmod php7.3
Considering dependency mpm_prefork for php7.3:
Considering conflict mpm_event for mpm_prefork:
Considering conflict mpm_worker for mpm_prefork:
Module mpm_prefork already enabled
Considering conflict php5 for php7.3:
Enabling module php7.3.
To activate the new configuration, you need to run:
  service apache2 restart

service apache2 restart
[ ok ] Restarting Apache httpd web server: apache2.

The pages now display correctly. I have only to remember the password and then I can carry on seeing what WordPress is about!

Geoff

#227 Re: Other Issues » Upgrading laptop from ASCII to Beowulf » 2019-03-25 14:16:56

I forgot to mention that I use LXDM as the login manager. I have tried some of the others but prefer this one. It has the advantage over at least one of the alternatives that you can shutdown or reboot from LXDM, which came in useful when I was having some difficulty shutting down when logged in (now sorted).
It is also clear how you select alternative desktops as it has a box at the bottom to choose from.

Geoff

#228 Other Issues » Upgrading laptop from ASCII to Beowulf » 2019-03-24 19:58:02

Geoff 42
Replies: 2

After trying Beowulf on a VM under Xen, I tried upgrading my Asus ZenBook UX305.
I started this with Debian Jessie and had upgraded over time to Deuan Jessie and then to ASCII.
Having replaced ascii with beowulf in /etc/apt/sources.list, then using apt-get update and apt-get dist-upgrade would not have upgraded the 3 elogind packages, so I tried aptitude full-upgrade and let it do its thing, which seems to have largely worked. I use LXQt with OpenBox and it needs a little bit of tweeking to get everything as I like it, but it seems to basically work.

However va-driver-all is marked as a broken package as it depends on mesa-va-drivers.

E: /var/cache/apt/archives/mesa-va-drivers_18.3.4-2_amd64.deb: trying to overwrite '/usr/lib/x86_64-linux-gnu/dri/nouveau_drv_video.so', which is also in package vdpau-va-driver:amd64 0.7.4-dmo6

Using the aptitude command which fsmithred supplied I looked for packages not in testing :-

aptitude search ~i -F"%p# %v# %t#"|grep -v testing
returns a lot of packages, (about 100 according to wc).

I then used Synaptic to carry out the tidying up.

I forced the version of upower to 0.99.10-1 which counts as a downgrade!
It also removes va-driver-all and moves to libupower-glib3, which gets rid of the broken package problem.
I downgraded vdpau-va-driver to the Beowulf version.

As I am using LXQt instead of LXDE, I removed the LXDE components.

I had consolekit installed as well as elogind, so remove consolekit,
which does some other sorting out of libpolkit stuff.
Initially the shutdown icon only logged me out, but with libpolkit sorted out, shutdown now works.

The old gksu from ASCII is still installed and works!

apt-get autoremove does some minor tidying up.

Where possible I forced versions of packages to the Beowulf ones.
Both Firefox-esr and Chromium have a version in Beowulf which appears to be older, so I forced the versions.
Also devuan-baseconf. Trying to force the version of that crashed Synaptic.
So I had to sort sources.list and removed the backports which don't seem ready yet!

I removed lots of packages which were not in Beowulf. The Beowulf stuff gets reported as "testing".
The others are reported as "now", which I guess means that it is not in the selected repositories.
Packages which I removed include :-
php7.0
ruby 2.1 & 2.2
leafpad
an older WordPress theme from ASCII
webkit-image-qt possibly used by the old version of Claws-mail and is now replaced by Dillo.
as well as numerous libraries and old versions of things, which had no dependencies.

After the version forcing and the removals I was left with gksu and some libraries on which it is dependent,
as well as keeping one old kernel, just in case it was needed.

$ aptitude search ~i -F"%p# %v# %t#"|grep -v testing | awk -e '{print $1,"   \t\t" $2,"\t" $3}' -
gksu    		2.0.2-9+b1 	now
libgksu2-0    		2.0.13~pre1-9+b1 	now
libgnome-keyring-common    		3.12.0-1 	now
libgnome-keyring0    		3.12.0-1+b2 	now
libgtop-2.0-10    		2.34.2-1 	now
libssl1.0.0    		1.0.2d-1 	now
linux-image-4.18.0-0.bpo.1-amd64    		4.18.6-1~bpo9+1 	now

Things are now largely running normally.

Geoff

#229 Re: Other Issues » Beowulf - new stuff » 2019-03-16 20:32:42

When I look in syslog I can see where it tries to start :-

grep have syslog
Mar 16 20:12:02 beowulf kernel: [    8.642908] audit: type=1400 audit(1552767120.069:5): apparmor="STATUS" operation="profile_load" profile="unconfined" name="/usr/sbin/haveged" pid=1517 comm="apparmor_parser"
Mar 16 20:12:02 beowulf haveged: haveged starting up

But it is not running, according to ps.

Geoff

#230 Re: Other Issues » Beowulf - new stuff » 2019-03-16 20:27:08

haveged is running ok in ASCII, but it is failing in a Beowulf VM.

In Beowulf, if I run haveged manually as root, then it keeps running :-

haveged -F -w 1024
haveged starting up

If I then ^C it and run :-

/etc/init.d/haveged start 
/etc/init.d/haveged status
[FAIL] haveged is not running ... failed!

But I noticed on the console :-

[  426.107690] kauditd_printk_skb: 5 callbacks suppressed
[  426.107691] audit: type=1400 audit(1552767537.537:17): apparmor="DENIED" operation="mknod" profile="/usr/sbin/haveged" name="/run/haveged.pid" pid=3276 comm="haveged" requested_mask="c" denied_mask="c" fsuid=0 ouid=0

so I am thinking that apparmor is stopping it from running and I have not yet got to grips with apparmor which seems to come as standard with Beowulf.

Geoff

#231 Re: Hardware & System Configuration » Automount problem » 2019-03-16 16:55:05

I wonder whether you mean using autofs. This way you can define the things you want mounted, but they are only actually mounted when you look at them.

Install autofs and then configure the auto.* files in /etc.

If I have a file server on my lan called sodium offering music then my auto.master might contain :-

/LAN    /etc/auto.lan  --ghost

the ghost flag is so that you can see the mount point all the time.

/etc/auto.lan might contain :-

music		-fstype=nfs,nolock	sodium:/volume1/music
# music		-fstype=cifs,user,uid=user1,gid=users,rw,suid,credentials=/etc/cifsmypwd	://sodium/music

I used to mount it via cifs but now I use nfs, so have commented out the cifs info.
My machine now only mounts /LAN/music if I go looking for it and unmounts it after a period of not using it.

Geoff

#232 Other Issues » Beowulf - elogind 421.1-1 - removes libsystemd0 » 2019-03-13 16:41:30

Geoff 42
Replies: 0

I see on DNG

https://lists.dyne.org/lurker/message/2 … 8e.en.html

that there is a new elogind and friends, which removes libsystemd0.

The only catch is that it needs testing ;-)

The only problem I have encountered so far is that

apt-get update
apt-get dist-upgrade

upgraded a lot of stuff but not the 3 elogind packages. After the large upgrade, I then went into synaptic and told it to upgrade these 3 packages. It did this correctly, with a note in the details that it was going to remove libsystemd0, but would do it anyway.

My Beowulf VM is now running without libsystemd0 and I can still fire up synaptic.

Geoff

#233 Re: Installation » Beowulf: Link sbin » 2019-02-18 15:49:39

Its a bit <tl;dr> and I was rather lazily hoping that someone would post the executive summary for us! I did try and work my way through it, but failed.

Geoff

#234 Re: Installation » Beowulf: Link sbin » 2019-02-18 09:52:19

There is some discussion about merging some of the bins on the mailing list. I think it is /usr/bin and /bin as well as /usr/sbin and /sbin

https://lists.dyne.org/lurker/message/2 … fe.en.html

Geoff

#235 Re: Other Issues » Copy, past urxvt-unicode in Devuan Ascii » 2019-02-15 08:56:36

I do not have a .Xresources file but it just works for me. I am using rxvt-unicode-256color under LXDE and LXQt and normally use left mouse for select and middle mouse click for paste, although I have just tested that right mouse also works for extending the selected area. Having read the above post, I checked that <ctl><alt>c and <ctl><alt>v do also work as described.

Geoff

#236 Re: Desktop and Multimedia » Software update notifications in XFCE » 2019-02-12 09:36:10

fsmithred wrote:

Last I checked, the man page for apt says not to use it in scripts. I don't know what bad things might happen if you do.

I think the point is that apt is meant to be easy to use, but the interface is not guaranteed to be stable, so if you want to put it into a script then the other programs, such as apt-get, are meant to have stable interfaces.

Geoff

#237 Re: Other Issues » Beowulf - Policykit » 2019-02-12 09:27:59

Thank you, I, too, now have it working.

My guess would be that pkexec/policykit is complaining about having a path in $TERM.

I opened the LXQt Configuration Centre and from there opened the Session Settings and in the Default Applications I set the Terminal Emulation to rxvt-unicode-256color (previously it was /usr/bin/rxvt). Then after I had logged out and back in, I was able to run Synaptic and Gparted from the menu. I then removed policykit-1-gnome and after logging out and in, it still all works.

As you say I have lxqt-policykit installed and ps reports that /usr/bin/lxqt-policykit-agent is running.

Geoff

#238 Re: Other Issues » Beowulf - Policykit » 2019-02-11 15:26:25

After a recent upgrade I have had some problems with some authentication. For example, selecting synaptic from the menu fails to run. I see in the thread about pkexec https://dev1galaxy.org/viewtopic.php?id=2667 that it is recommended to install  policykit-1-gnome. I have now done this and it has not caused any problems and I can run synaptic from the command line as synaptic-pkexec. I am still unable to run synaptic from the menu (using LXQt). Looking in /var/log/auth.log I can see that pkexec is complaining :-

The value for environment variable TERM contains suscipious content

In my window, $TERM is set to rxvt-unicode-256color. This doesn't look suspicious to me, but I am not pkexec! I wonder what it is picking up and what is wrong with it.
I have installed gparted and can run it from the command line but not from the menu, which gives the same error about TERM.

Geoff

#239 Re: Installation » Xen and the art of VM » 2019-02-08 16:37:38

Other USB devices

Having got a memory stick working I tried a portable DVD drive

usbip port
Imported USB devices
====================
Port 00: <Port in Use> at High Speed(480Mbps)
       LaCie, Ltd : unknown product (059f:0643)
       1-1 -> usbip://192.168.42.9:3240/1-2
           -> remote bus/dev 001/005

This worked and I was able to play an audio CD through VLC, once I had added myself to the group "cdrom". "eject" also worked.

Later I tried to put a video DVD in... Unfortunately this hung Dom0!

The DVD was mounted in Dom0 and then an info request came from DomU and then an attach request. This attachment seemed to complete and then there was a "general protection fault: 0000 [#1] SMP NOPTI" followed by a traceback from usbip-host on Dom0 and then the kernel went into a loop of tracebacks. Time to hit the power switch...

I spotted a USB bluetooth dongle for £1 in a charity shop, so I bought it. My Dom0 machine hasn't got bluetooth installed, but I attached it through to the Beowulf DomU and bluetoothctl show returns some plauseable looking data. I have yet to work out how to get it to actually do anything useful, despite having read man bluetoothctl!

bluetoothctl
Agent registered
[bluetooth]# help

is more informative, but bluetoothctl seems to have to be run as root to actually report anything.

[bluetooth]# discoverable on
Failed to set discoverable on: org.bluez.Error.Failed

The Bluetooth dongle could be seen, but I was not sure what to do with it. Connman was also able to see the dongle.

The next thing I tried was an external USB disk drive, a Samsung M3 Portable, which is USB 3 and, I recall 2TB. Unfortunately this did not work. I seemed that it was possible to connect it through to DomU but then it disconnects with the error :-

[ 5286.237272] usb usb2-port1: Cannot enable. Maybe the USB cable is bad?

USBIP seems to work in the general case, but some more extreme uses cause it problems. I think that it was problems with USBIP rather than Xen.

Geoff

#240 Re: Hardware & System Configuration » ssd optimization on devuan » 2019-02-05 09:49:12

Yes, that is the way to do it. There is a comment in that file, that this causes /tmp to be mounted earlier than /etc/fstab is read.

Geoff

#241 Re: Desktop and Multimedia » Software update notifications in XFCE » 2019-02-05 09:46:21

Although I use LXQt much of the time, it does mean that I have found lxqt-sudo. A quick check suggests that you could install it without pulling in too much else. I have been using this in Beowulf as a replacement for gksu.

Geoff

#242 Re: Hardware & System Configuration » ssd optimization on devuan » 2019-02-04 19:10:34

You can configure /tmp as a tmpfs in /etc/default/tmpfs.

Geoff

#243 Re: Installation » Xen and the art of VM » 2019-02-03 15:57:37

USB

The next thing I wanted to try was using USB devices in DomU.

The Xen's documentation on USB PassThrough
https://wiki.xenproject.org/wiki/Xen_USB_Passthrough
seems to hint that it doesn't work and if you are not able to use PCI PassThrough to give an entire USB hub to DomU then you could use client/server software to pass the USB control over the network connection, mentioning USBIP. I was unable to get the Xen USB PassThrough to work.

Some of the USBIP documentation seems to be a bit out of date.
https://sourceforge.net/p/usbip/git-win … ace/README

man usbip
usbip help

but with a bit of trial and error it works quite well. It is not restricted to Xen or VMs and should work between real machines as well.

On Dom0 (the host) install usbip:-

apt-get install usbip

modprobe usbip-core
modprobe usbip-host
usbipd -D
usbip list -l
 - busid 2-1 (046d:c52b)
   Logitech, Inc. : Unifying Receiver (046d:c52b)

 - busid 2-2 (0781:5151)
   SanDisk Corp. : Cruzer Micro Flash Drive (0781:5151)

usbip bind --busid 2-2
usbip: info: bind device on busid 2-2: complete

then on beowulf (DomU - the client) install usbip

apt-get install usbip

modprobe usbip-core
modprobe vhci_hcd

this only seems to work with the numeric address of the server

usbip list -r 192.168.42.9
Exportable USB devices
======================
 - 192.168.42.9
        2-2: SanDisk Corp. : Cruzer Micro Flash Drive (0781:5151)
           : /sys/devices/pci0000:00/0000:00:14.0/usb2/2-2
           : (Defined at Interface level) (00/00/00)

usbip attach -r 192.168.42.9 -b 2-2

I was then asked for authentication as Spacefm popped up. There was some confusion with one window complaining about another authentication in progress. I think this is a conflict between gvfs and Spacefm both trying to mount the device, with gvfs winning. I dismissed that one and authenticated as me and Spacefm had the usb memory stick mounted via gvfsd-fuse plus the 2 partitions displayed. I could unmount the gvfs in Spacefm and then unmount sda1 and sda2 (my 'real' disk is xvda2)

usbip port
Imported USB devices
====================
Port 00: <Port in Use> at High Speed(480Mbps)
       SanDisk Corp. : Cruzer Micro Flash Drive (0781:5151)
       1-1 -> usbip://192.168.42.9:3240/2-2
           -> remote bus/dev 002/005

usbip detach -p 0
usbip: info: Port 0 is now detached!

I have now removed gvfs and installed udevil. There is now no conflict over authentication. I can now mount and unmount the partitions at will in Spacefm.

Using usbip attach and detach seems to be like plugging and unplugging the memory stick.

usbip attach -r 192.168.42.9 -b 2-2
lsusb
Bus 016 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 015 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 014 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 013 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 012 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 011 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 009 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 010 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 008 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 007 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 006 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 005 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 002: ID 0781:5151 SanDisk Corp. Cruzer Micro Flash Drive
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub

usbip detach -p 0
usbip: info: Port 0 is now detached!

Back on Dom0 (the server) :-

usbip unbind -b 2-2
usbip: info: unbind device on busid 2-2: complete

and the device is again available on Dom0.

I have only tested this with a memory stick so far, but will try some other devices.

Geoff

#244 Re: Other Issues » Beowulf - Policykit » 2019-02-01 14:17:46

Whoops, I have just spotted that there is a newer version of the library libupower-glib3 with a version to match the newer upower.

I have now moved to the newer versions by using force-version in Synaptic, which refers to it as a downgrade!

Geoff

#245 Re: Other Issues » Beowulf - Policykit » 2019-02-01 14:14:19

Thank you for those commands. All of my Policy kit stuff is from testing.

Looking at where all the packages come from, with fsmithred's command, nearly all are from testing, with just a few being labelled "now".  If I grep for not testing then I find :-

aptitude search ~i -F"%p# %v# %t#"|grep -v testing
clearlooks-phenix-darkpurpy-theme 7.0.2-1+devuan2.0 now
darkpurpy-icon-theme 1.0 now
desktop-base 1:2.0.3 now
devuan-baseconf 0.6.4+devuan2.3 now
gnome-icon-theme-extras 3.12.0-1 now
libupower-glib1 1:0.9.23-2+devuan1.3 now
upower 1:0.9.23-2+devuan1.3 now

These must have been pulled in from ASCII when I was installing the earlier policykit stuff.

I believe that Golinux is working on the theme for Beowulf.

There is a newer version of upower in testing but the version number is 0.99.9-3 and there is no matching version of the library. I recall that there was a problem as to which version number would be taken as higher if there is a "1:" in front.

Geoff

#246 Other Issues » Beowulf - Policykit » 2019-02-01 10:36:00

Geoff 42
Replies: 13

Following the announcement about updated versions of policykit and friends having been put in Beowulf for testing purposes
https://lists.dyne.org/lurker/message/2 … 2b.en.html
I have now removed the pinning to the old ASCII versions and upgraded to these new testing versions. I believe that everything in my Beowulf VM is now from the Beowulf repositories. Is there an easy way to test this?

The initial report is that the basic stuff still seems to work, but KatolaZ has asked that the stuff round the edges is tested...

Geoff

#247 Re: Other Issues » Beowulf - new stuff » 2019-02-01 10:21:53

Another entry in daemon.log is for haveged. From /etc/init.d/haveged the description is :-

haveged uses HAVEGE (HArdware Volatile Entropy Gathering and Expansion)
 to maintain a pool of random bytes used to fill /dev/random whenever necessary.

This daemon appears to start but I cannot see it running and there are no error messages.

/etc/init.d/haveged status
[FAIL] haveged is not running ... failed!

Geoff

#248 Other Issues » Beowulf - new stuff » 2019-01-31 16:54:02

Geoff 42
Replies: 5

Having installed Beowulf, I notice that there are some new daemons some of which I don't recognise. These include :-

/usr/sbin/dundee
/usr/sbin/ofonod
/usr/sbin/connmand
/usr/sbin/bluetoothd

and in my name :-

/usr/lib/geoclue-2.0/demos/agent
/usr/bin/qlipper

connman is the new network connection manager, although I don't have much to manage, just the wired ether.
bluetooth; my desktop does not have bluetooth hardware and so I can disable that.
dundee does not have a man page and I have not found anything about it yet on the net. /etc/init.d/dundee says that it is a Bluetooth DUN daemon, which I think is for DialUp Networking.
ofonod is an oFono mobile telephony daemon, according to "man ofonod". Some Ubuntu docs report that it is responsible for the mobile telephony part and bluetoothd uses it to place and answer phone calls.
https://wiki.ubuntu.com/DebuggingBluetooth

geoclue works out where you are, using WiFi, GPS, 3G etc. Brought in by LXQt.
qlipper is a clipboard history applet, brought in by LXQt.

dmesg also reports that AppArmor is enabled. There is some info at :-

https://wiki.debian.org/AppArmor/HowToUse

# aa-status
apparmor module is loaded.
11 profiles are loaded.
9 profiles are in enforce mode.
   /usr/bin/man
   /usr/lib/x86_64-linux-gnu/lightdm/lightdm-guest-session
   /usr/lib/x86_64-linux-gnu/lightdm/lightdm-guest-session//chromium
   /usr/sbin/haveged
   libreoffice-senddoc
   libreoffice-soffice//gpg
   libreoffice-xpdfimport
   man_filter
   man_groff
2 profiles are in complain mode.
   libreoffice-oopslash
   libreoffice-soffice
0 processes have profiles defined.
0 processes are in enforce mode.
0 processes are in complain mode.
0 processes are unconfined but have a profile defined.

Adding "Z" to "ps" shows the Labels. Most things are "unconfined", but if you run LibreOffice then
ps axjfZ includes :-

LABEL                            PPID   PID  PGID   SID TTY      TPGID STAT   UID   TIME COMMAND

unconfined                       2657  3111  3111  2657 pts/1     3111 R+    1026   0:00      \_ ps axjfZ
libreoffice-oopslash (complain)     1  3055  3054  3054 ?           -1 Sl    1026   0:00 /usr/lib/libreoffice/program/oosplash --calc
libreoffice-soffice (complain)   3055  3072  3054  3054 ?           -1 Sl    1026   0:01  \_ /usr/lib/libreoffice/program/soffice.bin --calc --splash-pipe=5

Geoff

#249 Re: Other Issues » Beowulf - eudev and group kvm » 2019-01-31 09:24:04

Yes, that sounds like it. I would also understand from the link you give that the eudev people will be fixing the problem.

Geoff

#250 Re: Other Issues » Beowulf - eudev and group kvm » 2019-01-30 16:57:40

The actual warning from dmesg is :-

[    4.106843] udevd[373]: starting version 3.2.7
[    4.171397] random: udevd: uninitialized urandom read (16 bytes read)
[    4.171916] random: udevd: uninitialized urandom read (16 bytes read)
[    4.171933] random: udevd: uninitialized urandom read (16 bytes read)
[    4.259442] udevd[373]: specified group 'kvm' unknown
[    4.480625] udevd[374]: starting eudev-3.2.7

Geoff

Board footer

Forum Software