I have a:

• cloud instance
• remote server
• Virtual Machine
• ...

of Ubuntu to which I don't have physical access and I need to resize mounted partitions like / that I cannot dismount!

All the tutorials / answers I can find on-line are telling me to boot a USB or a DVD or ... but I can't because I don't have physical access to Ubuntu!

What should I do??? Help!

Sometimes I like to watch VLC media on my TV connected to the HDMI port on my laptop and I now have to:

• Start VLC on the TV:
  • Positioned on the TV by CCSM.
  • The screen saver is being turned off by VLC.
• Close the laptop
• Turn off the monitor (If it's on)

because I like to sit in the dark while watching TV!

Has anyone solved the above already for themselves through a script?

The TV is always connected to HDMI and the monitor always to the VGA port if that makes things easier.

So I've downloaded the free fan-created Star Trek Continues Blue-Ray ISO for Episodes 1-3 and I'm trying to convert the FHD 1920x1080 to SD 960x540 to be able to watch on my phone:

ffmpeg -canvas_size 960x540 \
-i /media/STC_VOL1_BR/BDMV/STREAM/00000.m2ts \
-map_metadata -1 \
-c:v libx264 -map 0:0 -strict -2 -r 24 -s 960x540 \
-c:a libmp3lame -map 0:1 -b:a 384K \
-metadata:s:s:0 language=EN -c:s dvdsub -map 0:2 \
/media/Videos/Star\ Trek\ Continues/S01E01\ Pilgrim\ Of\ Eternity.mkv

but after a few seconds of watching the converted mkv, the subtitles become:

• unreadable due to overlapping
• are slightly big (no issue)
• shifted to the right and bottom of the screen

Full ffmpeg output here

What am I doing wrong?

_{Note 1: I've only ever dealt with srt subtitles and I know that dvdsubs are images, so that's why I thought the canvas_size parameter might be of help, but it makes no difference.
Note 2: Yeah, it's not really a pure Ubuntu OS question, but over on video.se not getting any attention, even with a bounty so trying here if anyone does any video conversion and has run into the same situation.}

So I'm manually deleting old kernels and did a dpkg --list 'linux-image-*' and I get:

Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
||/ Name           Version      Architecture Description
+++-==============-============-============-=================================
ii  linux-image-4. 4.10.0-42.46 amd64        Linux kernel image for version 4.
ii  linux-image-4. 4.13.0-31.34 amd64        Linux kernel image for version 4.
ii  linux-image-4. 4.15.0-38.41 amd64        Signed kernel image generic
ii  linux-image-4. 4.15.0-39.42 amd64        Signed kernel image generic
ii  linux-image-4. 4.15.0-42.45 amd64        Signed kernel image generic
ii  linux-image-4. 4.4.0-128.15 amd64        Linux kernel image for version 4.
ii  linux-image-4. 4.4.0-140.16 amd64        Linux kernel image for version 4.
ii  linux-image-ex 4.4.0-128.15 amd64        Linux kernel extra modules for ve
ii  linux-image-ex 4.4.0-140.16 amd64        Linux kernel extra modules for ve
ii  linux-image-ge 4.4.0.140.14 amd64        Generic Linux kernel image
ii  linux-image-ge 4.15.0.42.63 amd64        Generic Linux kernel image
un  linux-image-un <none>       <none>       (no description available)
un  linux-image-un <none>       <none>       (no description available)
un  linux-image-un <none>       <none>       (no description available)

So I dug a little deeper and found that dpkg-query --show 'linux-image-*' doesn't help me much:

linux-image-4.10.0-42-generic   4.10.0-42.46~16.04.1
linux-image-4.13.0-31-generic   4.13.0-31.34~16.04.1
linux-image-4.15.0-38-generic   4.15.0-38.41~16.04.1
linux-image-4.15.0-39-generic   4.15.0-39.42~16.04.1
linux-image-4.15.0-42-generic   4.15.0-42.45~16.04.1
linux-image-4.4.0-128-generic   4.4.0-128.154
linux-image-4.4.0-140-generic   4.4.0-140.166
linux-image-extra-4.4.0-128-generic 4.4.0-128.154
linux-image-extra-4.4.0-140-generic 4.4.0-140.166
linux-image-generic             4.4.0.140.146
linux-image-generic-hwe-16.04   4.15.0.42.63
linux-image-unsigned-4.15.0-38-generic
linux-image-unsigned-4.15.0-39-generic
linux-image-unsigned-4.15.0-42-generic

So where are those 3 coming from and how do I get rid of those 3??

Additional info:

dpkg --list 'linux-image-*' > /tmp/N0rbert.txt is here.

uname -r and ls /boot are here

So I'm at a friend and hooked up my external EXT4 HDD to his router and it recognises it and he can connect his entire family to it and everything works fine.

However, on my Ubuntu machine I can:

• Go to my File Manager (Nemo)
• File Connect to Server

and that works fine!

but I cannot:

sudo mount -t smbfs //192.168.1.1/Fab-EXT4 /media/MichelNAS/
mount: unknown filesystem type 'smbfs'

sudo mount -t cifs //192.168.1.1/Fab-EXT4 /media/MichelNAS/
Password for root@//192.168.1.1/Fab-EXT4: 

sudo mount -t cifs -o guest //192.168.1.1/Fab-EXT4 /media/MichelNAS/
mount error(112): Host is down
Refer to the mount.cifs(8) manual page (e.g. man mount.cifs)

sudo mount -t cifs //192.168.1.1/Fab-EXT4 -o user=,password= /media/MichelNAS/ 
mount error(112): Host is down
Refer to the mount.cifs(8) manual page (e.g. man mount.cifs)

sudo mount -t cifs -o rw,noperm,sec=none //192.168.1.1/Fab-EXT4 /media/MichelNAS/
mount error(112): Host is down
Refer to the mount.cifs(8) manual page (e.g. man mount.cifs)

However:

smbclient --list=192.168.1.1 --no-pass

works perfectly and has as output:

WARNING: The "syslog" option is deprecated
Domain=[HOME] OS=[Unix] Server=[Samba 3.0.24]

    Sharename       Type      Comment
    ---------       ----      -------
    IPC$            IPC       IPC Service (mymodem Sharing Files)
    Fab-EXT4        Disk      Share Fab-EXT4 directory
Domain=[HOME] OS=[Unix] Server=[Samba 3.0.24]

    Server               Comment
    ---------            -------
    MYMODEM              mymodem Sharing Files

    Workgroup            Master
    ---------            -------
    HOME                 

So I'm obviously missing something completely stupid here, but I'm stumped as to why I cannot mount this share from the command line???

So when I type:

grep --recTab

I get:

grep --recursive

but

(and that's a big but) I have defined a few aliases:

alias grep-blue='GREP_COLOR='\''1;34'\'' grep --color=always'
alias grep-red='GREP_COLOR='\''1;31'\'' grep --color=always'
alias grep-yellow='GREP_COLOR='\''1;33'\'' grep --color=always'

which allow me to do funky stuff like this:

gsettings list-recursively | grep-yellow settings | grep-blue power | grep-red true

which gives me: However in my aliases, there is no prediction for the command-line parameters, so typing:

grep-blue --recTab

doesn't do anything!

and I had a look at:

How do I set up bash completion for command arguments?
How to create script with auto-complete?

but those seem to be a bit outdated and I can't seem to figure it out, because I did find that /usr/share/bash-completion contains:

# makeinfo and texi2dvi are defined elsewhere.
complete -F _longopt a2ps awk base64 bash bc bison cat colordiff cp csplit \
    cut date df diff dir du enscript env expand fmt fold gperf \
    grep grub head indent irb ld ldd less ln ls m4 md5sum mkdir mkfifo mknod \

so I added grep-blue grep-cyan grep-green grep-grey grep-magenta grep-red grep-white grep-yellow \ next to grep but that doesn't seem to help neither!

What am I missing here?

OK, here the premise: Customer is a Windows shop and their Ubuntu VMs always use 100% of their allocated memory (as per design) so they show up in "Red" on the memory usage charts. As we don't have access inside the VMs, the memory usage is measured externally by VMWare and Googling didn't give me any more insight as what I want to do is stupid.

It's also political:  I've told them already that this is by design, the machines do not use much swap and removing the Ubuntu VMs from the report is not feasible, nor is setting the memory reporting limit to 100% and monitoring swap usage only as some sales guy accepted these KPIs in the contract, but unfortunately that's the info I need to be able to raise a change request with their external provider for these VMs, so posting a question here.

Is there a way of limiting the Kernel's memory manager to use only 75% of available RAM under normal circumstances and only use the remaining 25% when it's really needed? (E.g. for Disk caching)

_{I'd like the memory still to be available when it's really needed, but I'll take any answer...}

What I tried already:

• Allocating a huge RAM disk and not using it: it's still allocated RAM... :-(

I have a bunch of live ISO files like GParted, CloneZilla, Boot-Repair, Ubuntu 14.04 32 bit, ... downloaded in /opt and on my old BIOS machine, I used to be able to just add them to the grub menu and boot them. :-)

I now have an UEFI machine and all the information I can find are for BIOS machines.

What are the steps to boot these ISO files from grub?

_{I'm just I'm sick and tired of "burning" these to a USB stick all the time as I can never find the stick I need whereas my computer itself if much harder to loose in the total chaos around me here...)} :-(

When I log in to a TTY I get the following nice text:

Ubuntu 16.04.03 LTS
Log-in: username
Password:
Last log-in date/time
Welcome to Ubuntu 16.04.03 LTS

 * Documentation:
 * Management:
 * Support:

9 packages to update
4 updates are security updates

I would like the last 2 lines repeated every time I open a GUI terminal for all users in the adm group (but will settle for having the entire thing displayed each and every time I open a GUI Terminal for all users)

I tried the obvious:

fab-root@fab-ux-predator:~
$ cd /etc/update-motd.d/
fab-root@fab-ux-predator:/etc/update-motd.d
$ ./90-updates-available 
fab-root@fab-ux-predator:/etc/update-motd.d
$ cat /var/lib/update-notifier/updates-available
cat: /var/lib/update-notifier/updates-available: Permission denied
fab-root@fab-ux-predator:/etc/update-motd.d
$

What am I missing?

P.S. Obviously I can chmod o+r the file, but how future-proof would that be?

I did some digging around and found that the automatic clean-up for kernels is in /etc/kernel/postinst.d/apt-auto-removal and the file states:

# In the common case this results in two kernels saved (booted into the
# second-latest kernel, we install the latest kernel in an upgrade), but
# can save up to four. Kernel refers here to a distinct release, which can
# potentially be installed in multiple flavours counting as one kernel.

but that script is way above my head and I cannot easily find how autoclean anything but the last 4 kernels.

I'd like to know how to:

1. take a backup
2. do a restore

of all the MBRs and all the PTs of all my disks to a file so I can easily back them up using rsync or a comparable program.

I'm well aware of the difference between symbolic and hard links. I'm also aware of the dangers of creating hard links to directories.

Nevertheless, I'm insisting on creating a hard link to a directory and all of its subdirectories (on the same partition).

However, the following does not work:

root@fab-ux:/home/fab-user/Public
# ln --directory ../Documents/CV/ CV
ln: failed to create hard link ‘CV’ => ‘../Documents/CV/’: Operation not permitted

Is there a kernel/File System (Running on EXT4) parameter that will allow me to accomplish this foolishness?

_{Yes, I can mount bind, can hard link individual files through a cron job, ... But the point is that I want both the convenience of physically only one directory and all its files and the convenience of only uploading certain hard-linked directories in my Public directory to the cloud and the cloud software detects symbolic links and doesn't support them and most important of all: I'm running Ubuntu! Not OSX nor Windows: I want the system to do what I want!}

Data

• I want operator users on this machine to mount their own cifs shares
• The sudoers file already contains the /bin/mount -t cifs //*/* /media/* -o username=* command for all operators
• I want the users to mount a cifs share through a script typing the password only once, not twice.
• The sudo password and the cifs password are identical.

What I already have

This script works:

#!/bin/bash
sudo 'mount -t cifs //192.168.1.1/home /media/$USER/home -o username=$USER'

...but it requires the users to type the same password twice!

• Once for sudo
• Once for the mount itself

This would also work:

#!/bin/bash
echo -n Password: 
read -s szPassword
echo $szPassword | sudo -S sh -c 'echo $szPassword | mount -t cifs //192.168.1.1/home /media/$USER/home -o username=$USER'

...but this would require me to allow all operator users to be able to sudo sh (major security problem)

Question

How to mount a cifs share in bash¹ without putting sh in the sudoers file nor creating a permanent/temporary file???

_{Note 1: no python, perl, C, Go, ... please?
Note 2: I know I can just remove the password through the sudoers file, but I'm trying to tighten security, not loosen it, without giving up convenience...}

When I plug my BQ Aquaris Ubuntu phone into my laptop running 14.04.3 LTS and shut down the laptop while the phone is charging the phone keeps charging...

Once I unplug the phone however, it doesn't charge any more after I plug it back into any of the the USB ports of the shut-down computer.

How can I continue charging my phone's battery off my laptop's battery?

_{I know: it's a bad idea as it'll drain the laptop battery, but this is an emergency and that's what I need right now! :P ;-)}

I'm a bit of a tinkerer and changed my most commonly used application icons to have my own steam-punk theme.

The only application whose icon I couldn't get to change was qBitTorrent.

The .desktop files are only present in /usr/share/applications to make them available to all users and ~/.local/share/applications/ has been verified not to contain any desktop files.

Can anyone shed a light on why this is happening? (And give some direction how to change that icon?)

For the ones out there who don't know about SteamPunk: I want to change the ugly blue icon... ;-)

Some more info:

$ lsattr /usr/share/applications/qBittorrent.desktop
-------------e-- /usr/share/applications/qBittorrent.desktop

$ lsattr /usr/share/pixmaps/Steampunk-Vlc-Audio-Player-Icon-Mkii.png 
-------------e-- /usr/share/pixmaps/Steampunk-Vlc-Audio-Player-Icon-Mkii.png

Now that I've got qTox running myself (with a little bit of help from @JacobVlijm) and no one has given me any ideas for qBittorrent, I'm adding a bounty...

Note:

I'm not looking to change the icon using the ~/.local/share/applications/ override but for all users in /usr/share/applications/ or any other system-wide setting!

On an Ubuntu 14.04.2 LTS the following command:

adduser fu

and then entering a password bar and just pressing Enter on all other questions, creates a user that allows me to log on and has the following characteristics in /etc/passwd:

fu:x:1002:1002:,,,:/home/fu:/bin/bash

However, doing a:

useradd fu --password bar --groups adm,cdrom,sudo,dip,plugdev,lpadmin --shell /bin/bash --create-home

creates in /etc/passwd

fu:x:1002:1002::/home/fu:/bin/bash

and does not allow me to log on and adding --user-group does not help neither...

As I need useradd (for a docker script) how do I proceed?

P.S. I've had a look at /usr/sbin/adduser already, but as it's perl it's not obvious to me which parameters to add to adduser to have the same functionality as useradd

This is not a question on how to install Ubuntu or how to use manual partitioning during installation but is a question on access rights and how not to let them bite you in the lower abdomen... ;-)

This question will not help you how to make partitions on your disk, but will tell you what to do in step 4 of the installation process, depending on your personal use case, so choose one of the following:

• I'm a beginning Windows and/or beginning Ubuntu user
• I know Windows well but Ubuntu is new / I know Ubuntu well, but Windows is new.
• I'm an expert at both Ubuntu and Windows!

Note: The above questions can also be used with multiple disks, but the examples in the answers are given for one disk.

No, this is not a comparison of back-up tools, nor is it a solicitation on What to backup or Why and only a little bit about Where to backup and When to do it.

This is basically a strategy question: what, where and when all together. "How" is not really relevant: there are tons of back-up programs out there and they all do the same basic thing: make copies of data.

But the real question is: are these back-up programs making copies of the data important to you? How should you install&run Ubuntu while safeguarding your data?

There is no "one size fits all" when it comes to computers, Ubuntu versions or a backup strategy, therefore I've split this up into 5 basic user types with each their own answer:

1. I don't have a computer!
2. My computer doesn't contain my life...
3. My computer contains my life!
4. My computer IS my life! (or you're running in UEFI mode!)

And as the above does not fit everyone, one for the impatient:

5. I want the fastest possible Restore!

So click one of the above links and go to your user type!

P.S. If you're running a server farm and you've read this question so far, I'm sorry: this is about a single PC, not a network, nor a server farm.

Given: I'm using export $(dbus-launch) in my /etc/bash.bashrc file as I'm normally running under a non-admin user and need to su - admin to do anything dangerous.

Normal behaviour: Go to dash, open gedit. The top menu is visible.

Abnormal behaviour Go to terminal, type: gedit. Then the top menu is invisible. (Problem persists even when sending to bg)

Additional info: I noticed when I do a gksudo gedit, that the menu appears below the menu bar of the window instead of the top menu…

More additional info: I just found the culprit: X11-forwarding. The "abnormal" behaviour stops when I remove export $(dbus-launch) from my bash.bashrc file.

Does anyone have a clue how to bring abnormal back to normal? (keeping x11-forwarding?) I've looked everywhere, but except for a bug on the preferences not showing, I found no solutions to this problem…