You are not logged in.
Thanks for the input!
@UnixRocks, @steve_v
Then tigerVNC and x2go it is for the next tries.
@nixer
Thanks for the detailed description, Xephyr sounds very interesting. I need time to get used to the concept, currently the nested / recursive X-server creates a knot in my head. As you mentioned, specifying the correct DISPLAY sounds like challange.
All three suggestions can be used with ssh, which is great.
PS:
Yesterday, while trying to theme lightdm ... .
The arch wiki mentioned a test mode lightdm --test-mode --debug, which works using xephyr. lightdm.conf has a VNC section and there is the installble lightdm-remote-session-x2go debian package. First I thought these components were addressing the server-side function, but the x2go module seems to be client related. Anyway.
USB to RCA audio & video (...) pic here: https://drive.google.com/file/d/1j7DbXk … share_link
I would guess the shown cable is an analog input, not an output.
It is made for digitisinging (recording) analog TV signals consisting of composit-video and audio, e.g. from video-tape-recorders (VHS/BetaMax/video8/whtever) or old video-cameras with TV-out.
it doesn't seem the image was resized (...), the restoring of the image file still results error.
I did not know that the size of the image file would be the same, sorry.
However. The file system in the image file has become smaller. The last 7 GB are present, but unused by now - df will tell.
$ sudo mount -o loop HP.shrinkme.img /mnt
$ dfThe error at restore using dd should not matter here. If you really want to be sure and write it again:
$ sudo ntfsclone -f -o /dev/sdb1 HP.shrinkme.imgThe file system should be in a mountable condition. Probably it isn't ... .
Did you try to mount /dev/sdb1 ?
If you want back the shrinked 7GB HDD space, you still need to:
$ sudo ntfsresize -x /dev/sdb1I deleted and recreated the windows partition at same location, not sure whether this leads to the partition change.
HDD situation:
There is the "master boot record" (mbr) and "partition start sector 2048" from the new installation, and the file system from the old installation (different mbr and "partition start sector 63") to fit the image size to the HDD space. I'm not aware of all implicit issues.
The disk itself will likely not boot. grub form the other disk may or may not be able to help ($ sudo update-grub).
Does the first version of the win7-install-cd the start-at-sector-63 partitioning? Think so, but not 100% sure. Probably, if you have such a install-cd, it may recreate the partition as before.
I'm not used to read chkdsk output.
Or just use the new installation and copy back all the files form the mounted image.
Hello,
I'm asking for suggestions for remote desktop access (the server component). What are you using?
ssh, vnc, rdesktop is basically all I know. "X-application with display on a different X-host" can be an option. Maybe there are other protocols commonly used. Something straight and simple would be nice.
The connection ability from anydesk / teamviewer to pass a router is not needed.
File transfer would be nice, but also not needed.
It may be a virtual session or mirror of the current display.
If there is a real killer application, it could be wayland too.
Sorry, but I need a client for windows too (work; can not change that).
xrdp: Currently used. The keymap issue is only one thing. Usage feels not very responsive, even in the same network. Moving a browser window ... raises +20 years old memories.
vnc: occasionally used, but good memories. Small and simple to setup. What would it be? tigervnc, tightvnc, x11vnc, ... ?
Thanks and regards.
@guuml.dev1: Thanks for the reply (yes, long ago). There are "ä,ö,ü,ß". [AltGr] is onyl working, if xrdp uses an appropriate keymap too.
Update - a cleaner way to set a default keymap for xrdp:
Initially loaded is /etc/xrdp/km-00000000.ini. The file is not existing, so a link or copy can be made without interfering the installation. The issue occurred again on a new PC.
In my case:
$ sudo cp -a /etc/xrdp/km-00000407.ini /etc/xrdp/km-00000000.iniAfter logging in, it is still a "us" or "en" keymap, but switching to "de" now fully works.
$ setxkbmap deFinally a suggestion:
The image file is working (mountable) and the target device is a bit smaller the image itself.
The solution is simple and straight forward: Resize the image to fit the device.
Get some information:
ntfsresize -i HP.qcow2ntfsresize works directly on the device or image file, a copy is recommended. The last command expands the filesystem to partiton size.
cp HP.qcow2 HP.shrinkme.img
ntfsresize -s 97G HP.shrinkme.img
sudo dd if=HP.shrinkme.img of=/dev/sda1 bs=10M
sudo ntfsresize -x /dev/sda1And it should be done.
Alternatively:
Grow /dev/sda1 to fit HP.qcow2.
Thanks for posting the file size.
Testing:
Just dd'ed a small 100M partition using bs=13M to have a "fractional block count". Sizes of image-file and device are equal. Also tried dd with "bs=15M" and "bs=512", and pure cat. All image files are identical. On daedalus and on chimaera. No errors at all during creation and restore.
The main question is now:
Why is the image-file (104864062464 bytes) bigger than the partiton (104862842880 bytes)? How is this possible?
The partititon must have been shrinked after creating the backup image.
I don't see any other possibility.
EDIT:
fdisk -lbefore the reinstalltion of windows would be informative.
Older (windows-)partition schema start at sector 63 (not sector 2048) and there is a gap of 398 sectors (or 0.2 MB) between sda1 and sda2. This is 1.2 MB potential addional space for sda1. The image file is 1.1 MB bigger than sda1.
Just thinking about the error messages - not a solution, to many assuptions! And probably unreadeable anyway.
tl;dr: Summery of the text below is: The last 5MB were (might) not written back to /dev/sdb1.
dd error:
Sizes of partition and disk-image:
The partition size of /dev/sdb1 by fdisk output: 204810240 Sectors * 512 Byte/Sector = 104862842880 Byte
The partition size of /dev/sdb1 divided in blocks of 10M is "10000.5" (echo 204810240*512/1024/1024/10 | bc -l). Block count in dd is not supposed to be a fractional number!Assumed behavior of dd during image creation: It read 10001 whole blocks. EDIT: WRONG at this point - the rest are follow-up errors!
Resulting image-file size then is: 10001 * 10M = 104868085760 byte (or 5MB bigger than /dev/sdb1). Please confirm or report the actual size of HP.qcow2.
This would cause the error when restoring.
dd parameters can be tuned to restore without an error:
(half block size and doubled block count result in integer numbers and the size of /dev/sdb1)
dd if=/media/erdos/easystore/Backup/HP.qcow2 of=/dev/sdb1 bs=5M count=20001 status=progressUnclear behaviour of dd in case of fractional blocks:
dd error: "10000+0 records out" may mean the last "half" block is not written at all (5MB less than size of /dev/sdb1)
dd error: "104862842880 bytes (105 GB, 98 GiB) copied" which is the size of /dev/sdb1.
On my system, I would try re-write only the last 5MB of the partition:
This would be fast compared to rewriting the whole 100GB. But I'm not 100% sure about the offset!
dd if=/media/erdos/easystore/Backup/HP.qcow2 of=/dev/sdb1 bs=5M seek=20000 skip=20000 count=1PS: Warning
All numbers may be wrong.
Don't do anything of the above you don't understand.
Mistakes incorporating dd can be very very harmful.
Did you use the appropriate "virtualbox-guest-additions-iso"?
It contains some graphic drivers for the guest-OS (and other stuff). Not sure whether this will cure "crappy graphics".
@zapper
Just curious: What happend in January 2021? Can't remember anything special by now.
Then ... What is the business model behind that campagne?
'partition' and 'drive' seem a bit mixed-up. To see what you have:
$ lsblkYou got me completely wrong! I'll try again.
The main point is: Would you pay for having more search options?
API access means you can use (in this case) the google seach engine directly in your own program or web side. This allows more search parameters and costs real money. No idea how much they are asking for, nor how much I would want to pay for that. Yandex, Bing and others are offering the same services. This has nothing to with profiling or using their web side directly.
On youtube, they doubled the amount of ads, along with the offer to have it ad-free for 10 bugs a month. Same technique may apply for the search.
Beeing naked in front of google and hoping they will deliver better results then is not an option for me. Google has become as evil as its compeditors. At least they delverd some useful and working software (the revolutionary search in the early 2000, maps, earth, and more I forgot).
What if you use (buy) an API key for google custom search?
Could be a way to have more search options. Or is it the same search, just the results are submitted differently? EDIT: This is most likely wrong.
Advanced search is canceled everywhere. It's a shame.
Not sure what is usable in your case.
Maybe remapping the whole num-block to portugues characters? This transforms the 105-keyboard to 88-key with special characters.
Or what about using an onscreen keyboard additionally?
The architecture of an Knoppix-stick (or iso) is quite interesting in that regard.
The root file system is on a compressed ISO image - 4.5GB for the DVD-image - read only file system is the point.
If thee is space on the USB-stick, it is possible to have a home directory for files and optionally the overlay-fs can be stored (default is in the RAM until reboot). So can have security updates or even install additional programs.
So: You could combine a read-only-root-fs without encryption with an encrypted overlay-fs for system updates.
(Overlay: Ralph posted yesterday about an overlay file system here https://dev1galaxy.org/viewtopic.php?pid=41386#p41386 Awesome! I was totally unaware of that and am still flashed).
Nice background image.
Can't say I'm using TDE, but really really appreciate the return of kpackage with it.
As far as I remember when I installed wine a while ago, when you just specify # apt install wine you just get wine64, the additional architecture does not matter. i386 is pre-requisite to do an # apt install wine32.
Interesting, you archived what I wanted at the time, but could not get.
Sorry, I did not ment to spread confuision.
EDIT:
Tested on Chimaera, it depends on 'recommended install'.
# apt install wine --no-install-recommendsdoes not install wine32, while
# apt install wine --install-recommends
# apt install wine64 --install-recommendsdoes, if i386 architecture is added.
Choosing to start the new HDD from the BIOS boot menu produces a "Welcome to GRUB!" message and a "grub_rescue" prompt - at which point I'm stuck.
I suspect that I need to find a way of either unlocking my existing encrypted disk first, to allow the GRUB installer to see it, or editing GRUB settings to recognise encrypted disks?
This should have worked.
The boot selection method via BIOS-boot-menue is independant from any installation on other HDD's, therefore no need to edit foreign grub configurations.
I probably would temporarily disconnect the Ubuntu-disk (physically), and then try to boot / chroot / rescue / reinstall.
PS:
(((the OP knows, but just to name it - the BIOS-boot-menue: Power-on the PC, it beeps, press once [F12_in_my_case] and the HDD selection appears.)))
As far as I remember
# dpkg --add-architecture i386
# apt install wine installed both, wine64 and wine32.
Indeet, it was impossible to get only wine64 without wine32 (trying to get away without i386-architecture).
EDIT: caution! My memory seems to trick me!
EDIT2: the above is true with apt's defaults or namely # apt install wine --install-recommends
Get black screens after grub too.
Newer Intel graphics (11-something?) seem a bit "picky" in what they display and what not. One feature is to not support BIOS boot anymore (black screen on non-efi-boot). As far as I remember, temporary using an exiisting PCIe graphics card showed boot messages.
Another thing is, when switching to a text-console, both screens showin the same content. No idea, whether this is as it should be.
Did you try to copy the URL form the network-portal-login-page visible in Chimaera's Firefox and re-use it in Daedalus?
Downloading Librewolf as we speak....
( https://librewolf.net/installation/debian/ )
For sound with alsa you probably need the "AppImage" - not the version form the debian repository.
Not sure why this is, but it was the case some months ago and did not re-check since then.
Thanks for the input, guys.
Probably it's not as bad as I thought in the first place. De-deprecation seems somewhat likely and gnu-which was included.
Still don't like it when "they" mess around with the basic tool set.
I think the NewInBookworm page just hasn't been updated
Best solution (EDIT:), but bad public relation.