Danatela's questions

I'm trying to detect and run partitions on top of jMicron fakeRAID, which is discovered by dmraid. How far I'm stuck with making this automatically during boot.

hopungo@hopungo-pc:~$ sudo kpartx -l /dev/mapper/jmicron_GRAID 
jmicron_GRAID1 : 0 7813722112 /dev/mapper/jmicron_GRAID 2048

Here is a good known solution from year 2010, but it is outdated in 2023 (probably, after 22.04). You simply will not find /etc/udev/rules.d/60-kpartx.rules. And if you make your own, it doesn't work.

I also tried mdadm, with the same result: I need a working way to run it on system start.

Another option is to use rc.local script, but it doesn't work too and also looks a bit crutchy. Is there any canonical way to run kpartx on start since 22.04?

I'm trying to improve this answer. As you can guess, add-apt-repository doesn't work through proxy, and accepted answer doesn't work for me. I tried:

• Setting and preserving local variables

  export http_proxy=http://proxy:port
  export https_proxy=$http_proxy
  sudo -E
  

  which just does nothing

• editing /etc/environment has same effect
• editing sudoers' file has same effect besides graphical apps cannot access mir

The mentioned answer helps, but doesn't fix the problem, because add-apt-repository cannot be simply modified to use proxy and keyserver port 80 every time. Lurking more I discovered that real problem is gpg configuration, which is badly documented. I just couldn't realize how to set http-proxy option or to honor-system-proxy permanently.

Trying to set option honor-http-proxy I wrote ~/.gnupg/dirmng.conf with such contents, but it didn't help:

honor-http-proxy

Also, I had no such problem before, in Ubuntu 8.04 — 16.04. Yep, the problem may be not in the proxy config, but in the keyserver port. If so, I'd like to set keyserver to use port 80 by default.

Any ideas are welcome.

I know this sounds ridiculous at first glance, but I started X session from root with

# X :1 vt08 &

So I guess sudo stop lightdm would be overkill. How can I stop that instance of X server without stopping main X session?

During configuration of one project, I got the error:

checking for JSON... no
no
configure: error: 
You must have JSON-GLib >= 0.14.0 development headers installed to build.

make: *** [config.status] Error 1

The configure script contains such strings:

$ cat configure | grep json-glib
    { { $as_echo "$as_me:${as_lineno-$LINENO}: \$PKG_CONFIG --exists --print-errors \"json-glib-1.0 >= 0.14.0\""; } >&5
  ($PKG_CONFIG --exists --print-errors "json-glib-1.0 >= 0.14.0") 2>&5
  pkg_cv_JSON_CFLAGS=`$PKG_CONFIG --cflags "json-glib-1.0 >= 0.14.0" 2>/dev/null`
    { { $as_echo "$as_me:${as_lineno-$LINENO}: \$PKG_CONFIG --exists --print-errors \"json-glib-1.0 >= 0.14.0\""; } >&5
  ($PKG_CONFIG --exists --print-errors "json-glib-1.0 >= 0.14.0") 2>&5
  pkg_cv_JSON_LIBS=`$PKG_CONFIG --libs "json-glib-1.0 >= 0.14.0" 2>/dev/null`
            JSON_PKG_ERRORS=`$PKG_CONFIG --short-errors --print-errors --cflags --libs "json-glib-1.0 >= 0.14.0" 2>&1`
            JSON_PKG_ERRORS=`$PKG_CONFIG --print-errors --cflags --libs "json-glib-1.0 >= 0.14.0" 2>&1`

So I assumed I need package json-glib-1.0 but it seems to be installed:

$ apt-cache policy json-glib-1.0
libjson-glib-1.0-common:
  Installed: 0.16.2-1ubuntu1
  Candidate: 0.16.2-1ubuntu1
  Version table:
 *** 0.16.2-1ubuntu1 0
        500 http://ru.archive.ubuntu.com/ubuntu/ trusty/main amd64 Packages
        100 /var/lib/dpkg/status
libjson-glib-1.0-0-dbg:
  Installed: (none)
  Candidate: 0.16.2-1ubuntu1
  Version table:
     0.16.2-1ubuntu1 0
        500 http://ru.archive.ubuntu.com/ubuntu/ trusty/main amd64 Packages
gir1.0-json-glib-1.0:
  Installed: (none)
  Candidate: (none)
  Version table:
gir1.2-json-glib-1.0:
  Installed: (none)
  Candidate: (none)
  Version table:
libjson-glib-1.0-0:
  Installed: 0.16.2-1ubuntu1
  Candidate: 0.16.2-1ubuntu1
  Version table:
 *** 0.16.2-1ubuntu1 0
        500 http://ru.archive.ubuntu.com/ubuntu/ trusty/main amd64 Packages
        100 /var/lib/dpkg/status

Does anyone know which package should I install? In the case it does matter, the source is Pidgin 3.0 development branch. Everything I found in Google is only a message about changing json-glib to 0.14 version.

As raaz suggested below, I downloaded, converted and installed RPM via alien. But the configure script produces the same error, even with the newest version of package.

$ dpkg -L json-glib-devel | grep include
/usr/include
/usr/include/json-glib-1.0
/usr/include/json-glib-1.0/json-glib
/usr/include/json-glib-1.0/json-glib/json-gvariant.h
/usr/include/json-glib-1.0/json-glib/json-builder.h
/usr/include/json-glib-1.0/json-glib/json-gobject.h
/usr/include/json-glib-1.0/json-glib/json-version.h
/usr/include/json-glib-1.0/json-glib/json-parser.h
/usr/include/json-glib-1.0/json-glib/json-reader.h
/usr/include/json-glib-1.0/json-glib/json-path.h
/usr/include/json-glib-1.0/json-glib/json-enum-types.h
/usr/include/json-glib-1.0/json-glib/json-glib.h
/usr/include/json-glib-1.0/json-glib/json-generator.h
/usr/include/json-glib-1.0/json-glib/json-types.h

I need this for 3 purposes:

1. For some reason, Ubuntu 13.04 on my home computer did not want to upgrade to 13.10. It said "No new releases found".

2. And some Ubuntu derivatives (such as Mint) do not use update-manager at all.

3. Also, in manual mode it is possible though not recommended to skip releases :P

I have strange problem. My Wallch icon is huge. See the screenshot:

I removed XFCE settings and reordered items a little. How can I restore normal Wallch icon size?

Update:

My /usr/share/applications/wallch-nautilus.desktop file:

[Desktop Entry]
Version=4
Name=Wallch
Comment=Change desktop wallpapers automatically
Exec=/usr/bin/wallch %U
Icon=wallch
Terminal=false
Type=Application
Categories=Utility;Application;
MimeType=inode/directory;

Actions=Change_Wallpaper;Start;

[Desktop Action Change_Wallpaper]
Name=Change Wallpaper Once
Exec=/usr/bin/wallch --change
TargetEnvironmet=Unity

[Desktop Action Start]
Name=Start/Stop Wallpapers
Exec=/usr/bin/wallch --start
TargetEnvironment=Unity

And the /usr/share/pixmaps/wallch.png:

I have read the Persistant File System vs Installation onto USB stick and realized I have my own question.

The problem is: when you do sudo apt-get update && sudo apt-get dist-upgrade from USB stick, it becomes unusable because it installs new kernel and doesn't properly update boot configuration. It boots to initramfs and stucks there. My only option is to reboot and recreate it again loosing my changes and time.

I know I can use UCK to update the ISO which I write onto that USB but it is easier to update inside that USB and make some changes to boot files.

So, how do I properly update once created USB stick with persistent file?

I made RAID 0 by mdadm'ing partition located on dmraid VIA VT6421 and Western Digital hdd. Installed Ubuntu on that RAID. Now, when I boot, I get (initramfs) ash prompt.

There, I have to write

mdadm --stop /dev/md1
mdadm --assemble --scan

And hit Ctrl+D to resume. Then Ubuntu loads normally.

I'd like to change that behavior to save my time. So, I need to change some script located in /scripts directory of initial ramdisk. But everything I know about initramfs is that it is gzipped. I looked into update-initramfs but didn't find the clue yet. Also, I'm afraid doing something there unless I know what I'm doing.

How do I change my initramfs script?

UPD: Now I know I should

gunzip -c initrd.img-3.13.0-24-generic | cpio -id

Then make my changes and recreate it with

find . | cpio -o | gzip -c >/boot/initrd.new.img-3.13.0-24-generic

But I still don't know which scripts I should edit.

UPD2: sudo parted -l

$ sudo parted -l
Model: JetFlash Transcend 32GB (scsi)
Disk /dev/sda: 32,1GB
Sector size (logical/physical): 512B/512B
Partition Table: msdos

Number  Start  End     Size    Type     File system  Flags
 1      131kB  32,1GB  32,1GB  primary  fat32        boot, lba


Model: WD My Passport 074A (scsi)
Disk /dev/sdb: 1000GB
Sector size (logical/physical): 512B/512B
Partition Table: msdos

Number  Start   End     Size    Type     File system  Flags
 1      1049kB  1000GB  1000GB  primary  ntfs


Model: ATA WDC WD1600JS-22M (scsi)
Disk /dev/sdc: 160GB
Sector size (logical/physical): 512B/512B
Partition Table: msdos

Number  Start   End    Size   Type     File system     Flags
 1      1049kB  160GB  160GB  primary  linux-swap(v1)


Error: Can't have a partition outside the disk!                           

Model: ATA WDC WD5000AAKS-0 (scsi)
Disk /dev/sde: 500GB
Sector size (logical/physical): 512B/512B
Partition Table: msdos

Number  Start  End  Size  Type  File system  Flags


Error: /dev/mapper/via_ecedbfaihb6: unrecognised disk label               

Model: Linux device-mapper (linear) (dm)
Disk /dev/mapper/via_ecedbfaihb5: 524MB
Sector size (logical/physical): 512B/512B
Partition Table: loop

Number  Start  End    Size   File system  Flags
 1      0,00B  524MB  524MB  ext4


Error: Can't have a partition outside the disk!                           

Model: Linux device-mapper (linear) (dm)
Disk /dev/mapper/via_ecedbfaihb2: 167GB
Sector size (logical/physical): 512B/512B
Partition Table: loop

Number  Start  End    Size   File system  Flags
 1      0,00B  167GB  167GB  ntfs


Model: Linux device-mapper (linear) (dm)
Disk /dev/mapper/via_ecedbfaihb1: 367MB
Sector size (logical/physical): 512B/512B
Partition Table: loop

Number  Start  End    Size   File system  Flags
 1      0,00B  367MB  367MB  ntfs


Error: Can't have a partition outside the disk!                           

Model: Linux Software RAID Array (md)
Disk /dev/md1: 320GB
Sector size (logical/physical): 512B/512B
Partition Table: msdos

Number  Start   End     Size    Type     File system     Flags
 1      1049kB  6494MB  6493MB  primary  linux-swap(v1)
 2      6494MB  27,5GB  21,0GB  primary  ext4
 3      27,5GB  320GB   293GB   primary  ext4

Transcription:

• /dev/sda & /dev/sdb each of 500GB is fakeraid named via_ecedbfaihb,
• Win8.1 reserved partition on via_ecedbfaihb1 (390MB),
• Win 8.1 system partition on via_ecedbfaihb2 (160GB),
• extended partition via_ecedbfaihb3,
• linux boot partition via_ecedbfaihb5 (500MB),
• mdadm 1st partition via_ecedbfaihb6 (159GB),
• data partition via_ecedbfaihb7 with the rest of space
• /dev/sdc1 mdadm second partition (159GB),
• /dev/md1 made of via_ecedbfaihb6 and sdc1,
• /dev/md1p1 swap (6GB),
• /dev/md1p2 extended,
• /dev/md1p5 root (20GB),
• /dev/md1p6 home (~300GB)

AFAIK alias is usually used to make 'shortcut' for long command. But what to do if I want to alias one command to another?

I need this because I replaced gnome-screensaver with xscreensaver, so the lock screen button does not work.

How to alias gnome-screensaver-command -l to be resolved into xscreensaver-command -lock?

I'm trying to sign in to my Windows Live account through System Settings → Online Accounts. After entering my credentials I can see the page asking me to allow Ubuntu access the account. After it, I see the initial page and account does not appear. That's what I see in the console:

(gnome-control-center:4915): account-plugin-WARNING **: AuthSession error: GDBus.Error:com.google.code.AccountsSSO.SingleSignOn.Error.NoConnection: Host  not found

(gnome-control-center:4915): credentials-cc-panel-CRITICAL **: cc-credentials-authorization-page.vala:182: Error completing auth session process: GDBus.Error:com.google.code.AccountsSSO.SingleSignOn.Error.NoConnection: Host  not found

The same happens with Google account. Can anyone explain what happens and how do I connect to my accounts?

When I was introduced with Ubuntu 8.04, the Yelp (Ubuntu documentation) was aware of man command: simply type man:<manpage> into the search string showed the desired man page in HTML look and feel. However, since release 13.04 this feature looks like has been removed.

Here is the screenshot of Yelp from 10.04:

Does anyone know how to turn it back?

I'm using XNeur to quickly switch between input languages (Russian and English). Since Ubuntu 13.10 implemented new method of switching keyboard layout it is not possible for this software to operate correctly now.

How can I revert back to legacy keyboard switching method?

Some screenshots:

I have recently installed Fedora in VirtualBox and noticed that they use Delta-RPM packages, which help dramatically reduce downloading sizes. Also I heard that Gentoo uses some delta packages. Would be glad to see such deltas in Ubuntu too.

Does anyone know about such delta-deb packages and how do I get to use them?

I know, I just can hit Super+A to see all installed apps in Ubuntu, but I need a command to list their names. The command

dpkg --get-selections | awk '{print $1}'

is also not an option because it shows all installed packages and it contains drivers, kernels and libraries.

I'd like to see the full How-To on how to use manual partitioning during Ubuntu installation. The existing guides (at least those I found here) cover only automatic part and leave untouched the manual part (or extremely short and contain no pictures).

I'd like to cover such situations:

• If you have blank disk
• If your disk contains Windows installed
• If your disk contains other systems:
• If it is GPT, RAID or LVM
• If you have a OEM-preinstalled copy of Windows 8

I built and installed my own version of package tora with Oracle InstantClient support using this guide. It has the same version as the package available from standart Ubuntu repository so Synaptic tells me it needs to be updated. I blocked it there, however, apt-get does not know about it and installs the default useless version everytime I do apt-get upgrade. How can I disable such behavior? I guess I should correct the build number of my package but I don't know where to hack sources. My tora version is 2.1.3-2build1.

There is RAID0 of 2 drives connected through Silicon Image 3132 SATA SoftRAID controller. Under Windows it was partitioned as one dynamic GPT-disk having 4 TB NTFS volume. There is a lot of music and movies on the drive. I'm trying to get him to be seen under Ubuntu as a single disk, not as 2 by 2 terabytes.

I tried to read it through dmraid, had no success, it is not displayed in /dev/mapper. Also tried to configure the kernel, but did not find anything suspicious, the driver for my controller was turned on. There is also a driver from the manufacturer, but it is only available for RHEL and SLES. Here it's reported that SoftRAID is supported by the kernel, but apparently not completely. If I thrust drives in the AMD controller, built into the motherboard, the drive is seen as a single, but the data is lost. I know about mdadm that it is able to ditch all the information on the disks. So, is it possible to somehow create an array without actually recording information on used drives and to make the system correctly identify sections on it later?

Information about the array:

/dev/sdf - Disk 0
/dev/sdg - Disk 1
Array type: Stripe
Chunk Size: 64KB

Also, a device /dev/md1 is created using command mknod /dev/md1 b 9 1