The Wiert Corner – irregular stream of stuff

Jeroen W. Pluimers on .NET, C#, Delphi, databases, and personal interests

  • My badges

    Mastodon

  • Twitter Updates

  • My Flickr Stream

    20140508-Delphi-2007--Project-Options--Cannot-Edit-Application-Title-HelpFile-Icon-Theming20140430-Fiddler-Filter-Actions-Button-Run-Filterset-now20140424-VMware-Fusion-6.0.3.-no-reclaimable
    More Photos

  • Pages

  • All categories

  • Enter your email address to subscribe to this blog and receive notifications of new posts by email.

    Join 1,839 other subscribers

Archive for the ‘Linux’ Category

« Previous Entries
Next Entries »

Some Linux mail solutions

Posted by jpluimers on 2018/09/07

Too bad most of them are very picky to the Linux distributions they run on.

–jeroen

Posted in *nix, Linux, openSuSE, Power User, SuSE Linux | Leave a Comment »

Linux: See Bandwidth Usage Per Process With Nethogs Tool – nixCraft

Posted by jpluimers on 2018/08/24

This tutorial explains how to find out network bandwidth usage per process in real time using nethogs tool under Linux operating systems.

Cool tool!

Source: [WayBack] Linux: See Bandwidth Usage Per Process With Nethogs Tool – nixCraft

An alternative is iftop – Wikipedia.

via:

–jeroen

Posted in *nix, *nix-tools, Linux, openSuSE, Power User, RedHat, SuSE Linux, Tumbleweed | Leave a Comment »

when bind named service hasn’t started after OpenSuSE Tumbleweed boots

Posted by jpluimers on 2018/08/20

A while ago, named would not start any more after I rebooted my Tumbleweed systems.

I had this behaviour on multiple systems, each installed quite a while ago and kept up-to-date with zypper dist-upgrade so it looked like a systematic issue.

Below are steps in researching the problem together with the helpful people on the IRC channel opensuse-factory.

Background reading for some of the commands: [WayBack] How To Use Systemctl to Manage Systemd Services and Units | DigitalOcean.

Both systemctl status named.service and systemctl status named would show the same output:

# systemctl status named
● named.service - LSB: Domain Name System (DNS) server, named
   Loaded: loaded (/etc/init.d/named; generated; vendor preset: disabled)
   Active: inactive (dead)
     Docs: man:systemd-sysv-generator(8)

Getting the log from events around a reboot would show a successful shutdown, but no start:

# journalctl --unit named --catalog --pager-end

Apr 28 13:19:27 laurel systemd[1]: Stopping LSB: Domain Name System (DNS) server, named...
-- Subject: Unit named.service has begun shutting down
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit named.service has begun shutting down.
Apr 28 13:19:28 laurel named[20360]: no longer listening on 192.168.124.27#53
Apr 28 13:19:28 laurel named[20360]: no longer listening on 192.168.124.27#53
Apr 28 13:19:32 laurel named[20360]: received control channel command 'stop'
Apr 28 13:19:32 laurel named[20360]: shutting down: flushing changes
Apr 28 13:19:32 laurel named[20360]: stopping command channel on 127.0.0.1#953
Apr 28 13:19:32 laurel named[20360]: no longer listening on ::#53
Apr 28 13:19:32 laurel named[20360]: no longer listening on ::#53
Apr 28 13:19:32 laurel named[20360]: no longer listening on 127.0.0.1#53
Apr 28 13:19:32 laurel named[20360]: no longer listening on 127.0.0.1#53
Apr 28 13:19:32 laurel named[20360]: exiting
Apr 28 13:19:34 laurel named[30705]: Shutting down name server BIND  waiting for named to shut down ..done
Apr 28 13:19:34 laurel systemd[1]: Stopped LSB: Domain Name System (DNS) server, named.
-- Subject: Unit named.service has finished shutting down
-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
-- 
-- Unit named.service has finished shutting down.

Similar results in these files:

  • /var/lib/named/log/general.log

28-Apr-2017 13:19:32.465 general: shutting down: flushing changes
28-Apr-2017 13:19:32.468 general: stopping command channel on 127.0.0.1#953
28-Apr-2017 13:19:32.622 general: exiting

  • /var/lib/named/log/named.log

28-Apr-2017 13:19:32.489 network: no longer listening on ::#53
28-Apr-2017 13:19:32.489 network: no longer listening on 127.0.0.1#53

With systemctl, I got this:

# systemctl is-enabled named
named.service is not a native service, redirecting to systemd-sysv-install.
Executing: /usr/lib/systemd/systemd-sysv-install is-enabled named
enabled
# systemctl is-active named
inactive
# systemctl is-failed named
inactive

After this, I was out of systemd and sysv knowledge, so I asked for help on the #openSUSE-factory IRC channel, where ismail was of great help.

Read the rest of this entry »

Posted in *nix, *nix-tools, bind-named, Linux, openSuSE, Power User, SuSE Linux, Tumbleweed | Leave a Comment »

Just I in case I need to port CombineApacheConfig.py to OpenSuSE properly

Posted by jpluimers on 2018/07/24

I came across a nice tool that combines httpd.conf files:

python CombineApacheConfig.py /etc/apache2/httpd.conf /tmp/apache2.combined.conf

In case I ever need to fully port it to OpenSuSE, I’ve put it in the gist below.

For now it works fine on OpenSuSE when used with the above command. I might make the default depend on the kind of nx it runs on.

via:

–jeroen

Read the rest of this entry »

Posted in *nix, *nix-tools, Apache2, Development, Linux, openSuSE, Power User, Python, Scripting, Software Development, SuSE Linux | Leave a Comment »

(35) Enabling New Hardware in U-Boot – Jon Mason, Broadcom Ltd. – YouTube

Posted by jpluimers on 2018/07/20

Lots of interesting information, especially these discussions where Alexander Graf chipped in:

The hand off of the boot loader to the actual OS payload. The payload and U-boot are in memory running at the same time. The payload can call back into U-boot through the uEFI API that U-boot implements so that the payload provides extra drivers enabling for instance a subsequent graphical stage (splash screen, menus, etc), more hardware access and so on. You even could pass ACPI tables through U-boot to the payload and help shooting yourself in the foot.

Important aspects for upstreaming:

  • keep commits short so they are easy to review
  • make sure patches are always rebaseable for each and every commit set (so it compiles throughout)
  • this tremendously helps doing a git bisect
  • it makes adding features that other parts depend on hard: you need to think on chicken & egg situations in advance

–jeroen

 

Posted in *nix, Development, Hardware Development, Linux, Power User, Software Development, U-Boot | Leave a Comment »

PowerShell on Mac OS X and other non-Windows systems

Posted by jpluimers on 2018/07/17

I wasn’t expecting it to be so easy to install PowerShell on Mac OS X:

brew install Caskroom/cask/powershell

In the background it executes this script: https://github.com/caskroom/homebrew-cask/blob/master/Casks/powershell.rb. which indirectly goes through the URL template https://github.com/PowerShell/PowerShell/releases/download/v#{version}/powershell-#{version}.pkg.

On other non-Windows systems, you have to go through GitHub yourself: https://github.com/powershell/PowerShell. The PowerShell team at Microsoft has many more repositories including the Win32-OpenSSH port which you can find through https://github.com/PowerShell.

At the time of writing, PowerShell was available for these platforms:

Platform Downloads How to Install
Windows 10 / Server 2016 (x64) .msi Instructions
Windows 8.1 / Server 2012 R2 (x64) .msi Instructions
Windows 7 (x64) .msi Instructions
Windows 7 (x86) .msi Instructions
Ubuntu 16.04 .deb Instructions
Ubuntu 14.04 .deb Instructions
CentOS 7 .rpm Instructions
OpenSUSE 42.1 .rpm Instructions
Arch Linux Instructions
Many Linux distributions .AppImage Instructions
macOS 10.11 .pkg Instructions
Docker Instructions

The first version I installed on Mac OS X was this: ==> Downloading https://github.com/PowerShell/PowerShell/releases/download/v6.0.0-alpha.17/powershell-6.0.0-alpha.17.pkg

By now I really hope it is out of Alpha state.

–jeroen

via:

Posted in *nix, Apple, CommandLine, Development, iMac, Linux, Mac, Mac OS X / OS X / MacOS, MacBook, MacBook Retina, MacBook-Air, MacBook-Pro, MacMini, openSuSE, Power User, PowerShell, PowerShell, Scripting, Software Development, SuSE Linux, Ubuntu | Leave a Comment »

btrfs free space. It’s complicated. Still.

Posted by jpluimers on 2018/07/09

Everytime a btrfs based volume runs out of space, I’m reminded of these:

There are a few scripts that help you assess quota usage. If you think they are wrong, then you need to btrfs quota rescan / which tells you that it started, but won’t tell when it’s finished (nor wil journalctl -xe a.k.a. journalctl --catalog --pager-end), but dmesg does:

# dmesg | grep qgroup
[ 316.608122] BTRFS info (device sda2): qgroup scan completed (inconsistency flag cleared)

For now I’ve this quick script to start investigation:

~/Versioned/btrfs-du/btrfs-du && df -h | grep "\/$\|^[^\/]" && btrfs quota rescan -s /

It assumes there is quota on the root (enable with btrfs quota enable /) and is based on my fork github.com/jpluimers/btrfs-du. The df will limit itself to the root (trailing / matched by \/$) or disks not mounted from / (matched by ^[^\/]).

–jeroen

References (not solutions):

Read the rest of this entry »

Posted in *nix, btrfs, File-Systems, Linux, openSuSE, Power User, SuSE Linux | Leave a Comment »

USB serial converters from FTDI – how to select them and get rid of fake chips

Posted by jpluimers on 2018/06/08

Be sure to read the comments at USB serial converters from FTDI are quite popular – Thomas Mueller (dummzeuch) – Google+ [WayBack] as it tells more about how to set genuine and fake FTDI chips apart (often before buying them, you cannot tell) and how to replace fake ones buy genuine ones.

Getting back to original means de-soldering fake chips, and soldering new genuine chips on the boards.

Fake chips:

  • have limited batches of the same serial (although there are ways to reprogram the serial, see links below)
    • connecting multiple adapters with the same serial causes trouble
  • are mangled by various FTDI drivers (either their PID is reset, or fake-data is inserted in the serial stream)
  • have problems operating at higher data rates

Note that the workarounds for these fake chips mean you cannot use more recent chips.

Links from Thomas’s post and other relevant links on the various kinds of genuine/fake and getting fake ones work again

Fake chips can cause you a lot of headaches – and time – sorting out communication problems: [WayBack] esptool-ck, esp8266, and FTDI Bug Hunting – vilimblog

Two years ago, this shoot-out (with results on github) had a great conclusion:

Buy either an adapter with a genuine FTDI chip, or one of the Silicon Labs CP2102 chips.

The FTDI chip is the only one attaining 3M baud rates.

Overview articles:

SiLabs chips seem to be the only without much trouble:

  • [WayBack] SiLabs CP210x USB Adapters For The Win – vilimblog
  • Review of a Generic USB RS 232/485/TTL Adapter (“Winners” branded) – YouTube:
    • Almost anything with FTDI/Prolific chips that comes from ebay/aliexpress should be assumed to be counterfeit – easy way to test this is to buy 2 and see if their serial numbers match. You don’t really want to waste your time with fake ones, you’ll end up spending more on those if you try to get them for as cheap as possible. Itead sells an adapter for 6.8 USD with genuine FT232RL chips, but they aren’t in as convenient dongle form factor, but might be worth more than your own time replacing fake chips with genuine ones.Although not mentioned here, CH340/1 should generally be avoided, while they work fine for short periods of time, for longer use they seem to be as unreliable as the counterfeit FTDI/Prolific chips on Linux.

      I haven’t seen or recognized a fake CP210x chip yet, so my guess would be that these are generally fine no matter where you get them. One big advantage of those is the integrated voltage 3.3V regulator, which can save some space when rolling your own boards.

FTDI related:

Genuine FTDI seems to be the easiest to find:

Prolific has similar issues, driver v3.4.25.218 still works with fake chips:

–jeroen

Related posts where I mentioned some of the trouble with FTDI chips:

Read the rest of this entry »

Posted in *nix, Development, Hardware Development, Hardware Interfacing, Legacy Ports: COM, Linux, openSuSE, Power User, SuSE Linux | Leave a Comment »

Hmm, latest Tumbleweed on Raspberry Pi 3 has an odd thing booting: Ansi termi…

Posted by jpluimers on 2018/05/25

Reminder to self: since the [Archive] ARM Cortex A53 problems are fixed, check out if this is also fixed:

[WayBack] Hmm, latest Tumbleweed on Raspberry Pi 3 has an odd thing booting: Ansi terminal characters aren’t being processed any more. – Jeroen Wiert Pluimers – Google+

Related: Git repository with fixed binaries for Tumbleweed on Raspberry Pi 3 – Bug 1084419 – Glibc update to 2.27 causes segfault during name resolution

–jeroen

Related as a different Pi died after a sudden power down:

Posted in *nix, Development, Hardware Development, Linux, openSuSE, Power User, Raspberry Pi, SuSE Linux, Tumbleweed | Leave a Comment »

Ubuntu: Fixing the myserious “Failed to stop apt-daily.timer: Connection timed out”

Posted by jpluimers on 2018/05/14

Ubuntu is the only Linux system I had that – after installing a text-mode console setup – gets itself in the below state with only running apt update and apt-get upgrade.

Preparing to unpack .../archives/apt_1.2.19_armhf.deb ...
Failed to stop apt-daily.timer: Connection timed out
See system logs and 'systemctl status apt-daily.timer' for details.
Failed to get load state of apt-daily.timer: Connection timed out
dpkg: warning: subprocess old pre-removal script returned error exit status 1

I could not find meaningful search results for the above thing, nor did systemctl status apt-daily.timer return anything better than

Failed to get properties: Failed to activate service 'org.freedesktop.systemd1': timed out

Heck, it doesn’t even reboot any more (no helpful search results either):

# reboot
Failed to start reboot.target: Failed to activate service 'org.freedesktop.systemd1': timed out
See system logs and 'systemctl status reboot.target' for details.
Failed to open /dev/initctl: No such device or address
Failed to talk to init daemon.

Nor did systemctl status reboot.target return anything better than

Failed to get properties: Failed to activate service 'org.freedesktop.systemd1': timed out

From the ubuntu-16.04-minimal-odroid-c1-20160817.img.xz base system, it only had these extra packages installed:

  • etckeeper
  • tmux
  • speedtest-cli
  • sendemail

Workaround found, but still unsure why this happened in the first place

Despite the workaround below, I still don’t know why I got into this situation and this leaves me with a very uncomfortable feeling.

Too bad some systems (like ODroid) only have Ubuntu based distributions as sorting out stuff like above took me way too much effort.

Read the rest of this entry »

Posted in *nix, Development, etckeeper, Hardware Development, Internet, Linux, Odroid, Power User, SpeedTest, Ubuntu | 2 Comments »

« Previous Entries
Next Entries »