Ludwig Schulze's questions

I would like to use skype on an old netbook with a 32bit-only atom processor (N270).

The skype linux client is only available for 64 bit linuxes. But they have a web app. However, their web app is only supported for the browsers "Edge" and "Chrome".

Edge is a windows browser and does not run on linux. Chrome is no longer available for 32 bit linux. But Chromium is. As far as I understand, Chromium is the same as Chrome except for some additional Google integration.

Skype web service actually starts with Chromium and lets me log into my account, I get much farther with it than with Firefox, which is immediately blocked by Skype web. However, I cannot participate in audio and video calls with chromium, the buttons for initiating them are greyed out, and a tool-tip says "Audio and video calls are not supported on your browser."

My question is can I make Chromium more appear like Chrome to circumvent this not-Chrome detection and hopefully use Skype on this otherwise unused old netbook (The processor is an Atom N270 and does not support 64bits).

I have low-latency kernels installed on my ubuntu notebook because I often process live audio in real time with low latency, and I find that I can achieve smaller sound card buffer sizes with the low-latency kernels than with the generic kernels.

Sometimes I have to use virtualbox on the same notebook. I find that I cannot start virtualbox virtual machines when running a low-latency kernel. The error messages are included at the end of the question. I need to reboot into the generic kernel in order to be able to run virtualbox VMs. I remember that I was able to run virtualbox VMs also with the low-latency kernels in the past. Is there a workaround? The error messages suggest to reinstall virtualbox-dkms and to run 'modprobe vboxdrv', however, the first suggestion does not help, and the second suggestion does not succeed.

Error messages:

Failed to open a session for the virtual machine 'VM name'. Details: The virtual machine 'VM name' has terminated unexpectedly during startup with exit code 1 (0x1).

Result Code: NS_ERROR_FAILURE (0x80004005) Component: MachineWrap Interface: IMachine {some GUID}

Another error pops up before acknowledging the first error:

VirtualBox - Error un suplibOsinit

The VirtualBox Linux kernel driver is either not loaded or not set up correctly. Please resintall virtualbox-dkms package and load the kernel module by executing

'modprobe vboxdrv'

as root.

[...]

where: suplibOsInit what: 3 VERR_VM_DRIVER_NOT_INSTALLED (-1908) - The support driver is not installed. On linux, open returned ENOENT.

Some packages list dependencies in their Depends: field like python:any or perl:any. What does the :any part mean?

Why would a package maintainer choose to list a dependency on python or perl in that way with :any?

In ubuntu 18.04 with default desktop, behaviour of logout/login has changed:

Previously on an Ubuntu system, when I find that I need to add my user id to some group, it was sufficient to

sudo adduser ludwig docker # adds me to group docker

and then I had to logout and login again to make the group change effective.

I notice that with Ubuntu 18.04, after adding the group and logging out and back in, the list of effective groups is still unchanged.

As a workaround I rebooted the system, which is inconvenient (requires making the right selection in grub and re-entering the disk encryption password).

1. Why is the behaviour now like this?
2. Can I do something else short of rebooting?

(I know I can ssh into localhost and get the correct groups in the ssh session only. This is also too inconvenient.)

eth0 is managed by the network manager icon and connects using dhcp when it detects a network cable.

However, very occasionally I want to connect to some special-purpose computers with static IP addresses using a direct ethernet connection (cross-over). Then I set

sudo ifconfig eth0 10.90.90.91

or

sudo ifconfig eth0:1 10.90.90.91

and can connect to the computer in question.

However, after a minute or so, that eth0 looses that IP address and my connections hang until I repeat the ifconfig step.

Right now I'm using a loop and repeat setting the IP address every 10 seconds.

But I wonder if I can tell the network manager (which I suspect is the culprit) somehow to, right now, not try to manage eth0 until I need it again to do it.

Edit: The question How do I set an extra IP for an existing interface? differs from mine. That question wants to perpetuate the setting across reboots. I'm not interested in that. I only want to have that fixed IP address for eth0 very occasionally. Usually the dhcp procedure is ok for me. Only sometimes I want it to back off.