Arseni Mourzenko's questions

I'm using Ubuntu 18.04.1 LTS. I want to disable network manager on an Ubuntu machine, because (1) I don't need it, (2) I prefer having hardcoded configuration, and (3) network manager regularly causes issues by changing the DHCP configuration.

I tried to follow the official documentation:

Stop network manager

sudo systemctl stop NetworkManager.service

Disable network manager (permanently) to avoid it restarting after a reboot

sudo systemctl disable NetworkManager.service

Despite this, the network manager is back again every time I reboot the machine.

How can I make it go away?

I noticed that /var/spool/nullmailer directory contains a weird trigger entity which is displayed in yellow and has a “|” suffix.

It doesn't seem to be a directory (at least I can't cd into it), and when I try to open it with vi, the console freezes.

What is that thing?

I'm trying to connect a Nexus 7 tablet (Android 6.0.1) from Ubuntu 14.04 LTS.

I started by following the instructions from this answer. When I execute the last command android-connect, it tells me:

Unable to open ~/.mtpz-data for reading, MTPZ disabled.Listing raw device(s)
Device 0 (VID=18d1 and PID=4ee1) is a Google Inc (for LG Electronics/Samsung) Nexus 4/5/7/10 (MTP).
   Found 1 device(s):
   Google Inc (for LG Electronics/Samsung): Nexus 4/5/7/10 (MTP) (18d1:4ee1) @ bus 2, dev 4
Attempting to connect device
Android device detected, assigning default bug flags
Error 1: Get Storage information failed.
Error 2: PTP Layer error 02fe: get_handles_recursively(): could not get object handles.
Error 2: Error 02fe: PTP: Protocol error, data expected
Listing File Information on Device with name: (NULL)
LIBMTP_Get_Storage() failed:-1

Searching for a solution led me to this answer, which tells that:

1. GVFS-MTP is required.

   In order to get the latest version of GVFS-MTP, I added, as instructed on another site, the corresponding PPA. Unfortunately, apt-get update terminates with:

   W: Failed to fetch http://ppa.launchpad.net/itachi-san/gvfs-mtp/ubuntu/dists/trusty/main/binary-amd64/Packages  404  Not Found
   W: Failed to fetch http://ppa.launchpad.net/itachi-san/gvfs-mtp/ubuntu/dists/trusty/main/binary-i386/Packages  404  Not Found
   E: Some index files failed to download. They have been ignored, or old ones used instead.
   

2. A bug in libmtp can be solved using a method explained in the author's blog, given that according to the author:

   The method outlined here is meant to be an experimental / quick fix for testing

   which is not particularly reassuring.

According to this other question, I have to:

• apt-get install mtpfs, which I already did when following the instructions from the first linked answer,

• Ensure that USB configuration is set to MTP. Did that, it was already set to MTP.

I also rebooted the PC and enabled and disabled USB debugging, unplugged/plugged the device and tried a different USB cable and different USB port, with no effect.

Is there something else I can try?

I noticed that when deploying a virtual machine while using a preseed, the host name specified in the preseed is ignored and the one from DNS server is used instead.

Detailed description

The Ubuntu Server machine is deployed using:

virt-install \
    --name custom-name \
    --initrd-inject /some-path/preseed.cfg \
    --extra-args "preseed/file=preseed.cfg" \
    ...

The preseed file contains, among others, the following lines (notice the custom-name value in the second line):

...
d-i netcfg/disable_dhcp boolean true
...
d-i netcfg/get_hostname string custom-name
d-i netcfg/get_domain string pelicandd.com
d-i netcfg/get_ipaddress string 192.168.1.35
...

On the other hand, DNS server contains both the A record:

demo    IN A       192.168.1.35

as well as the PTR record:

35      IN  PTR     demo.example.com.

Once the machine is created, it appears that /etc/hostname contains:

demo

and that /etc/hosts contains:

127.0.0.1       localhost
192.168.1.35    demo.example.com     demo

Question

I'm not particularly happy with the fact that the deployment process accesses the DNS server to get the information. Not only does it take time, but in some circumstances, the DNS server can be down (such as the case where it's the DNS server itself which is being deployed).

What should I specify in the preseed to prevent the installer from accessing DNS to retrieve information which is already available in the preseed?

In a very similar question, the accepted answer suggests to do:

dd if=/dev/zero of=/tmp/output bs=8k count=10k;

to measure the write speed of a hard disk.

My tests were inconclusive. I have an LVM-based software RAID 1 setup with two ordinary hard disks. When running the aforementioned command, it gives the following result:

10240+0 records in
10240+0 records out
83886080 bytes (84 MB) copied, 0.0784284 s, 1.1 GB/s

Not that I'm unhappy to have a RAID array which can store 1 GB per second, but I would still like to know:

• Why am I getting wrong results? Is it something to do with cache? RAID? LVM?

• How to get the actual file-level write performance?

When mounting an NFS directory by doing:

sudo mount 192.168.1.5:/home/shared /mnt/common

I get the following error:

mount: wrong fs type, bad option, bad superblock on 192.168.1.5:/home/shared, missing codepage or helper program, or other error (for several filesystems (e.g. nfs, cifs) you might need a /sbin/mount.<type> helper program)  
In some cases useful info is found in syslog - try dmesg | tail or so.

The kernel is up to date.

The question was already asked (for example here and here), but are either not answered or the answers are not helpful in my case.

What is wrong?

On a server machine, I have the /etc/network/interfaces file similar to:

# The loopback network interface
auto lo
iface lo inet loopback

# The primary network interface
auto eth0
iface eth0 inet static
    address 192.168.1.35
    netmask 255.255.252.0 
    network 192.168.0.0
    broadcast 192.168.3.255
    gateway 192.168.1.1 
    dns-nameservers 192.168.1.3 192.168.1.4
    dns-search example.com

When running resolvconf -u, the /etc/resolv.conf file becomes:

# Dynamic resolv.conf(5) file for glibc resolver(3) generated by resolvconf(8)
#     DO NOT EDIT THIS FILE BY HAND -- YOUR CHANGES WILL BE OVERWRITTEN
nameserver 192.168.1.3
nameserver 192.168.1.4
search pelicandd.com
# Dynamic resolv.conf(5) file for glibc resolver(3) generated by resolvconf(8)
#     DO NOT EDIT THIS FILE BY HAND -- YOUR CHANGES WILL BE OVERWRITTEN
nameserver 192.168.1.13

Why is there a nameserver 192.168.1.13?

According to this page, Gnome Terminal supports 24bit true color. This is consistent with Wikipedia.

When I run color-spaces.pl in Ubuntu 14.04 LTS with Gnome Terminal 3.6.2, the result is the following:

While there is no reference image, I imagine that those green strangely-shaped objects at the bottom indicate that the terminal is unable to print 24bit colors correctly.

How to enable 24bit support?

As I understand, desktop and server distributions of Ubuntu are very similar, except that some packages are included in one but not another.

I need to install Ubuntu Server from PXE. Since there is no way to install official Ubuntu Server from PXE, but is easy to do with the desktop variant, what do I need to do to change the list of packages (and only that) on an Ubuntu Desktop ISO in order to be able to have an Ubuntu Server variant?

I set up a PXE server which hosts both Ubuntu Desktop and Ubuntu Server.

• Ubuntu Desktop is installed successfully from PXE.

• Ubuntu Server seems to successfully load vmlinuz and initrd.gz, asks for the language, then the location, then the keyboard layout, and finally complains that it can't mount the CD-ROM:

The content of /var/lib/tftpboot/pxelinux.cfg/default is the following:

default ubuntu-installer/amd64/boot-screens/vesamenu.c32
menu title Ubuntu setup
label ubuntu-14.04-desktop-amd64
    menu label ubuntu-14.04-desktop-amd64
    kernel ubuntu-14.04-desktop-amd64/vmlinuz.efi
    append initrd=ubuntu-14.04-desktop-amd64/initrd.lz root=/dev/nfs boot=casper netboot=nfs nfsroot=192.168.1.41:/exports/ubuntu-14.04-desktop-amd64 splash --
label ubuntu-14.04-server-amd64
    menu label ubuntu-14.04-server-amd64
    kernel ubuntu-14.04-server-amd64/vmlinuz
    append initrd=ubuntu-14.04-server-amd64/initrd.gz root=/dev/nfs boot=install netboot=nfs nfsroot=192.168.1.41:/exports/ubuntu-14.04-server-amd64 splash --

What explains the fact that it requests the CD-ROM and how to avoid it?

I'm testing the creation of virtual machines with Ubuntu. The first virtual machine was created with the default settings and is called ubuntu:

virsh --connect qemu:///system list --all
 Id    Name                           State
----------------------------------------------------
 -     ubuntu                         shut off

I'm creating an additional virtual machine:

vmbuilder kvm ubuntu \
--suite saucy \
    --flavour virtual \
    --user demo \
    --pass demo \
    --arch amd64 \
    --dest /var/vm/demo1 \
    --part vmbuilder.partition \
    --domain demo1 \
    --ip 192.168.1.3 \
    --mask 255.255.255.0 \
    --gw 192.168.1.1 \
    --dns 192.168.1.2 \
    --libvirt qemu:///system \
    --addpkg linux-image-generic \
    --addpkg openssh-server

The command fails with the following error:

Domain ubuntu already exists at qemu:///system

Why is it sticking with ubuntu? Isn't it the purpose of --domain to set the name of the domain?

I'm trying to set up PXE in order to install Ubuntu 13.10 Server x64 through local network. So far, I can boot through PXE and, when using pxelinux.cfg/default provided by default, be able to install Ubuntu by downloading it from a mirror.

I want to be able to install Ubuntu from NFS. I edited pxelinux.cfg/default like this:

1  # D-I config version 2.0
2  default ubuntu-installer/amd64/boot-screens/vesamenu.c32
3  menu title Ubuntu setup
4  label ubuntu-13.10-server-amd64
5          menu label ubuntu-13.10-server-amd64
6          kernel 3.11.0.18-generic
7          append netboot=nfs initrd=install/initrd.gz nfsroot=192.168.1.75:/↵
   exports/ubuntu-13.10-server-amd64 ip=dhcp rw

• 192.168.1.75 is the IP of the server I use for DHCP, DNS and NFS,
• I got the kernel 3.11.0.18-generic by running uname -r,

• showmount -e 192.168.1.75 gives:

  Export list for 192.168.1.75:
/exports/ubuntu-13.10-server-amd64 192.168.1.0/24

  The directory contains the contents of the original ubuntu-13.10-server-amd64 image.

When booting from PXE, I get the original screen (see screenshots below), then another one. Pressing Tab shows the command:

3.11.0.18-generic netboot=nfs initrd=install/initrd.gz nfsroot=192.168.1.↵ 75:/exports/ubuntu-13.10-server-amd64 ip=dhcp rw

Pressing Enter does nothing.

I imagine there is something wrong with my changes of pxelinux.cfg/default.

What's wrong with it?

Is there a comprehensive documentation of the options to set?

Are there logs which can help understanding what is wrong?

I created a MAAS server and a node. When I boot (through PXE) the virtual machine corresponding to this node, it starts loading Ubuntu. A few minutes later, it is stuck in what seems to be an infinite loop, printing regularly the following message:

[ 239.617011] INFO: task touch:1060 blocked for more than 120 seconds.
[ 239.618857] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.

The same message (only the number at the beginning change) is repeated, over and over. I kept it running for probably more than half an hour, and the result is the same.

This results in “Failed tests” status for the concerned node.

The error appears both when using default installer and fast installer.

The file /var/log/maas/maas.log contains no errors.

Where can I gather more information about the cause of this issue?