Pasky’s Log

Wow, what a jargon-filled post title. Basically, we do a lot of our deep learning currently on the AWS EC2 cloud – but to use the GPU there with all the goodies (up to CuDNN that supports modern Theano’s batch normalization) is a surprisingly arduous process which you basically need to do manually, with a lot of trial and error and googling and hacking. This is awful, mind-boggling and I hate that everyone has to go through this. So, to fix this bad situation, I just released a community AMI that:

• …is based on Ubuntu 16.04 LTS (as opposed to 14.04)
• …comes with CUDA + CuDNN drivers and toolkit already set up to work on g2.2xlarge instances
• …has Theano and Keras preinstalled and preconfigured so that you can run the Keras ResNet model on a GPU right away (or anything else you desire)

To get started, just spin up a GPU (g2.2xlarge) instance from community AMI ami-f0bde196 (1604-cuda80-cudnn5110-theano-keras), ssh in as the ubuntu@ user and get going! No hassles. But of course, EC2 charges apply.

Edit (errata): Actually, there’s a bug – sorry about that! Out of the box, the nvidia kernel driver is not loaded properly on boot. I might update the AMI later, for now to fix it manually:

1. Edit /etc/modprobe.d/blacklist.conf (using for example sudo nano) and append the line blacklist nouveau to the end of that file
2. Run sudo update-initramfs -u
3. Reboot. Now, everything should finally work.

This AMI was created like this:

• The stock Ubuntu 16.04 LTS AMI
• NVIDIA driver 367.57 (older drivers do not support CUDA 8.0, while this is the last driver version to support the K520 GRID GPU used in AWS)
• To make the driver setup go through, the trick to install apt-get install linux-image-extra-`uname -r` per
• CUDA 8.0 and CuDNN 8.0 set up from the official though unannounced NVIDIA Debian packages by replaying the nvidia-docker recipes
• bashrc modified to include cuda in the path
• Theano and Keras from latest Git as of writing this blogpost (feel free to git pull and reinstall), and some auxiliary python-related etc. packages
• Theano configured to use GPU and Keras configured to use Theano (and the “th” image dim ordering rather than “tf” – this is currently non-default in Keras!)
• Example Keras deep learning models, even an elephant.jpg! Just run python resnet50.py
• Exercise: Install TensorFlow on the system as well, release your own AMI and post its id in the comments!
• Tip: Use nvidia-docker based containers to package your deep learning software; combine it with docker-machine to easily provision GPU instances in AWS and execute your models as needed. Using this for development is a hassle, though.

Enjoy!

We have our mice TV now streaming our colony of mus minutoides at the canonical URL http://mice.or.cz/ but it would be nice if you could watch them in your web browser (without flash) instead of having to open a media player for the purpose.

I gave that some serious prodding. We still use vlc with the same config as in the original post (mp4v codec + mpegts container). Our video source is an IP cam producing mp4v via rtsp and an important constraint is CPU usage as it runs on my many-purpose server (current setup consumes 10% of one CPU core). We’d like things to work in Debian’s chromium and iceweasel, primarily.

It seems that in the HTML5 world, you have these basic options:

• MP4/H264 in MP4 – this *does not work* with live streaming because you need to make sure the browser receives a correct header with metadata which normally occurs only at the top of the file; it might work with some horrible custom code hacks but nothing off-the-shelf
• VP80/VP90 in webm – this works, but encoding consumes between 150%-250% CPU! even with low bitrates; this may be okay for dedicated streaming servers but completely out of the question for me
• Theora in Ogg – this almost works, but the stream stutters every few seconds (or slips into endless buffering), making it pretty hard to watch; apparently some keyframes are lost and Theora homepage gives a caveat that Ogg encoding is broken in VLC; the CPU usage is about 30%, which would have been acceptable

That’s it for the stock video tag formats, apparently. There are two more alternatives:

• HTTP Live Stream (HLS) has no native support in browsers outside of mobile, might work with a hack like https://github.com/RReverser/mpegts but you may as well use MSE then
• Media Source Extensions (MSE) seem to allow basically implementing decoding custom containers (in javascript) for any codecs, which sounds hopeful if we’d just like to pass mp4v (or h264) through. The most popular such container is DASH, which seems to be all about fragmenting video to smaller HTTP requests with per-fragment bitrate negotiation, but still completely codec agnostic. Re Firefox, needs almost latest version. Media players support DASH too.

So far, the best looking courses seem to be:

• Media server nginx-rtmp-module (in my case with pull directive towards the ipcam’s rtsp) with mpeg-dash output and dash.js based webpage. I might have misunderstood something but it might actually just work (assuming that the bitrate negotiation could always end up just choosing the ipcam’s fixed bitrate; something very low is completely sufficient anyway).
• Debug libogg + libtheora to find out why it produces corrupted streams – have fun!

A couple of weeks ago, I have created my own fairly elaborate setup of the Conky system monitor. I have been wanting to fix up some of the weather display aspects, but I’m realistically not getting around to that anytime soon.

So, I have pushed it out to Github now.

My Firefox (or rather Iceweasel) is prone to constantly spinning and eating about 50-70% CPU on average when should be supposed to just sit idle. I tried to find the root cause, but Javascript profiler sees nothing and other forays didn’t end up with much (I discovered that spinning progress wheel when some tab is forever loading amounts to about 20% CPU, though). Not sure if always, but sometimes it spins within WebGLImageConverter::run() (no callgraph, sorry).

So in the end I decided to treat the symptoms instead. Cleaning my CPU fan helped with the noise, but the main problem is that running firefox brings my battery life to about 1/2 to 1/3. So one obvious solution would be to just stop the damn process when I don’t use it. I typically don’t do background downloads while on battery (or otherwise), so that means I want it stopped when the window is inactive – not in focus. This is surprisingly exotic idea, apparently, and not easy to do in most window managers.

I even tried switching to awesome or i3 window managers which should make this easy, but I’m psychologically not up to that; I think I’m too conservative, but I decided not to stick with that. I use the MATE desktop environment with marco window manager. Perhaps switching to sawfish would be a good option, but in the end I just decided to write a shell script that will periodically assess the situation and suspend or resume firefox as needed. Of course this introduces extra wakeups and ambient CPU load, but when powertop reports that my GoogleTalkPlugin (running at all times for whatever reason) wakes up 150 times per second, the powersaving situation on Linux is still too messy – so who cares?

Here goes the script, in the hope that it will be useful for someone else too. Run it in a terminal or backgrounded in your ~/.xprofile, it will stop the firefox process when out of focus for more than 10s and on battery, and resume it within a second when switching back. In practice, I found these timings completely acceptable so far, and didn’t notice any ill effects of constant STOP/CONT either.

firefox-suspender.sh:

#!/bin/bash
#
# firefox-suspender: Periodically check whether firefox is out of focus
# and STOP it in that case after a time delay; if in focus but stopped,
# send SIGCONT.
#
# (c) Petr Baudis <pasky@ucw.cz>  2014
# MIT licence if this is even copyrightable
 
loop_delay=1    # [s]
stop_delay=10   # [s]
 
last_in_focus=$(date +%s)
firefoxpid=
state=running
 
while true; do
  sleep $loop_delay
 
  # Get active window id
  window=$(xprop -root _NET_ACTIVE_WINDOW)
  window=${window#*# }
  # What kind of window is it?
  class=$(xprop -id "$window" WM_CLASS)
  # echo Active window $window, class $class
 
  if [[ "$class" =~ Navigator ]]; then
    # Firefox!  We know it is running.  Make sure we
    # have its pid and update the last seen date.
    # If we stopped it, resume again.
    if [ "$state" = stopped ]; then
      echo "$(date)  Resuming firefox @ $firefoxpid"
      if kill -CONT $firefoxpid; then
        state=running
      else
        firefoxpid=
      fi
    fi
    last_in_focus=$(date +%s)
    if [ -z "$firefoxpid" ]; then
      firefoxpid=$(pidof iceweasel)
    fi
    if [ -z "$firefoxpid" ]; then
      firefoxpid=$(pidof firefox)
    fi
 
    continue
  fi
 
  # Not Firefox!  If it's running, we are on battery and
  # it's been long enough, stop it now.
  if [ "$state" != running ]; then
    continue
  fi
 
  read battery </sys/class/power_supply/BAT0/status
  if [ "$battery" != Discharging ]; then
    continue
  fi
 
  if [ $(($(date +%s) - last_in_focus)) -ge $stop_delay ]; then
    echo "$(date)  Stopping firefox @ $firefoxpid"
    if ! kill -STOP $firefoxpid; then
      firefoxpid=
    fi
    state=stopped
  fi
done

An improved version of this script: https://github.com/mkoura/browser-suspender

I recently switched to xterm-256color in my konsole, but I found that vim looks exceedingly ugly, unfortunately. The colors were all washed out and difficult to read with reduced brightness. I decided to explore some alternative color schemes, including popular ones like solarized etc., but they just don’t work for me – I have the default color scheme burned into my mind and I really like its high contrast even though I can easily stare at it for 12+ hours in a row. It also works great even in adverse light coditions on a notebook.

In the end, I didn’t find any dark vim color scheme that would just look like its default 16-color color scheme. (Light background color scheme looks mostly the same in 16 and 256 colors by default.) So I had to create my own – 256like16.vim, drop it in ~/.vim/colors. You may want to edit yours to add some more bolds to make it look exactly like the 16-color scheme, but I ended up liking this one more, after all.

(You will need to install colorsupport.vim so that GUI color settings are used in the 256-color terminal. This particular script worked by far the best for me, and :ColorSchemeBrowse is also great when exploring schemes.)

Call me old-fashioned but I’m still using xterm on my desktop computer (where I use just fluxbox as my window manager) – it suits me just fine, but for one thing that I finally managed to solve. xterm by default ignores the clipboard, and none of the previously published solutions cut it for me, until now.

In X11, we have two commonly used selection buffers – PRIMARY and CLIPBOARD:

• PRIMARY is used when you simply highlight text in most applications, without pressing anything, and you can paste from it using the middle mouse button; it is of fleeting nature and used for quick cut’n’paste; and it doesn’t work well with all applications, e.g. libreoffice doesn’t put highlighted stuff there at least in some contexts and non-textarea HTML5 text edit widgets usually can’t handle the middle button for pasting.
• CLIPBOARD is used when you use ctrl-c or ctrl-v and can be used even with the evil applications above, but the problem is it’s not supported by xterm well!

In most terminal emulators, you can use the clipboard either using menus or shift-ctrl-c / shift-ctrl-v. However, in xterm, the best you can do is either…

• Make it use CLIPBOARD just instead of PRIMARY and in the same manner – the moment you select any text in xterm, it will plaster it over whatever else was in the CLIPBOARD before, without any explicit action. This sucks.
• Have a different set of bindings for selection to PRIMARY and CLIPBOARD. This is a lot better, but I’m out of modifiers since I use shift to cut’n’paste in terminal applications that use mouse themselves (e.g. elinks).

So, my solution is to bring in the shift-ctrl-c / shift-ctrl-v bindings! In your ~/.Xresources or ~/.Xdefaults, add

XTerm*VT100.*translations:      #override \
        Shift Ctrl C: select-end(CLIPBOARD, CUT_BUFFER0) \n\
        Shift Ctrl V: insert-selection(CLIPBOARD, CUT_BUFFER0)

(and don’t forget to xrdb ~/.Xresources afterwards).

Now, you can use shift-ctrl-v for pasting from CLIPBOARD, and almost use shift-ctrl-c for copying to clipboard. There is a catch – you must press shift-ctrl-c while you are still holding the mouse button, i.e you press left mouse button, drag your selection, then before releasing it, press shift-ctrl-c; thankfully, that can be done by one hand without too much cramping.

It’s a bit inconvenient because of this bug, and doesn’t quite work with left and right selection; maybe I will sometime get around to adding true clipboard support to xterm code, but I think this is good enough for me at this point. :)

I don’t send email from my notebook often, dealing with my correspondence on my server machine via ssh. When I need to do it, it’s usually when I’m sending Git patches or something like that. I didn’t meet much trouble with sending it directly, but SMTP servers of Debian-involved people are some of the most picky one can meet and I decided it’ll be best if I switch the exim4 on my notebook to smarthost mode where all mail is relayed via my main server.

So that should be trivial to do, right? Wrong, apparently. I figured I’d use SMTP auth, but it just seems mind-bogglingly complicated to configure if you don’t want to spend an evening on it. The client part is fairly easy (probably both on exim4 and postfix), but setting up postfix server to do SMTP auth (for just a single person) is really silly stuff. Maybe not so crazy if you use PAM / shadow for authentication, but that means that on my notebook, I’d have to store (in plaintext) my server password anyone could use to log in – no way. It seems I could switch to Dovecot and somehow pass it a simple password to use, but at that point my patience ran out and I just backed off a litle.

Why not just use ssh for smarthost SMTP transport? Authentication via ssh is something everyone understands nowadays, it does the best job there, no silly passwords involved and you can just pipe SMTP through it. You wouldn’t do that at in a company setting with Windows notebooks, but for a single geek, it seems ideal.

Someone already did set up ssh as exim transport, but that’s for exim3. So here follows a super-quick HOWTO to do this with exim4:

• Set up ssh key on client:

  sudo -u Debian-exim /bin/bash
  ssh-keygen # go with the default, and empty password, this will be used in an automated way
  ssh me@server.example.org # to fill up known_hosts; it will fail yet
  cat ~/.ssh/id_rsa.pub # this is my public key
  exit # ..the sudo
  

• Set up ssh key on server – paste the public key printed by the cat above to ~me/.ssh/authorized_keys and prepend command="nc -w1 localhost smtp",no-agent-forwarding,no-port-forwarding,no-X11-forwarding to the key line. This key can now be used only for mail relaying.
• Do dpkg-reconfigure exim4-config and configure smarthost mode. Also use it to find out whether you are using split or big configuration. You will also probably want to enable “mailname hiding”, otherwise your return-path will contain an unroutable address.
• Set up ssh transport in exim4 – add the following to the config file:

  ssh_pipe:
    debug_print = "T: ssh_pipe for smarthost delivery"
    driver = pipe
    path = "/bin:/usr/bin:/usr/local/bin"
    command = "ssh me@server.example.org nc -w1 localhost smtp"
    use_bsmtp
    message_prefix = "HELO mynotebook.example.org\r\n"
    delivery_date_add
    envelope_to_add
  

  (it would be nicer if we used the actual smarthost configuration option value and our notebook’s hostname instead of hardcoded strings, I guess).

• In the smarthost: section of the configuration file, replace transport = remote_smtp_smarthost with transport = ssh_pipe.
• /etc/init.d/exim4 reload and voilá, sending mail from anywhere should work now!

I *wish* setting up roaming SMTP nodes would be way easier nowadays and I wouldn’t have to eventually spend about 90 minutes on this stuff…

Inspired by the Debian CTTE deliberations on the new default init for Debian, I installed systemd on my notebook after tonight’s forced reboot and played with it a little.

(And I like it! I was very sceptical when hearing about systemd first, but after reading a lot of discussions and trying it myself, I find most of the problematic points either fixed already or a load of FUD. The immediate big selling point for me is actually journald, it and its integration with systemctl is really awesome. I’ll actually find systemd more useful on servers than desktops, I think.)

While it’s a nice exercise for anyone wanting to get familiar with systemd, I still decided to share a tidbit – service file that will make log entries show up on /dev/tty12. Many people run with rsyslogd set up for this, you’ll want to disable that (by default, all journal entries are forwarded to rsyslog). The advantage of showing journal entries instead is mainly color coding. :)

The file listing follows, or get it here.

# Simple systemd service that will show journal contents on /dev/tty12
# by running journalctl -af on it.
# Install by:
#  - Saving this as /etc/systemd/system/journal@tty12.service
#  - Running systemctl enable journal@tty12
#  - Running systemctl start journal@tty12
# journald can also log on console itself, but current Debian version won't
# show timestamps and color-coding.
# systemd is under LGPL2.1 etc, this is inspired by getty@.service.

[Unit]
Description=Journal tail on %I
Documentation=man:journalctl(1)
After=systemd-user-sessions.service plymouth-quit-wait.service systemd-journald.service
After=rc-local.service

# On systems without virtual consoles, don't start any getty. (Note
# that serial gettys are covered by serial-getty@.service, not this
# unit
ConditionPathExists=/dev/tty0

[Service]
# the VT is cleared by TTYVTDisallocate
ExecStart=/bin/sh -c "exec /bin/journalctl -af > /dev/%I 2> /dev/%I"
Type=idle
Restart=always
RestartSec=1
UtmpIdentifier=%I
TTYPath=/dev/%I
TTYReset=yes
TTYVHangup=yes
#TTYVTDisallocate=yes
TTYVTDisallocate=no
KillMode=process
IgnoreSIGPIPE=no

# Unset locale for the console getty since the console has problems
# displaying some internationalized messages.
Environment=LANG= LANGUAGE= LC_CTYPE= LC_NUMERIC= LC_TIME= LC_COLLATE= LC_MONETARY= LC_MESSAGES= LC_PAPER= LC_NAME= LC_ADDRESS= LC_TELEPHONE= LC_MEASUREMENT= LC_IDENTIFICATION=

[Install]
Alias=getty.target.wants/journal@tty12.service

(P.S.: Creating this service file – my very first one – took me 10 minutes total, including studying documentation and debugging two stupid mistakes I made.)

Edit (2019-10-19): New versions of journalctl check their stderr to decide whether to use colorize output – I have updated the recipe accordingly.

Pro zobrazování poloh dopadů meteosond na IRC jsem potřeboval v jednoduchém CSV formátu seznam souřadnic českých měst, ale ukázalo se, že je překvapivě obtížné něco takového získat. Sice existuje tabulka na jednom astronomickém webu, výběr tam zahrnutých obcí je ale docela divný, někde je místo obce jen její část, atd.

Nakonec jsem zvolil postup “udělej si sám”, a to kombinací seznamu na Wikipedii, Google Geocoding API a trochy XPath.

Seznam rozumné podmnožiny měst mohu získat třeba pomocí:

curl 'http://cs.wikipedia.org/w/index.php?title=Seznam_obc%C3%AD_s_roz%C5%A1%C3%AD%C5%99enou_p%C5%AFsobnost%C3%AD&action=edit' |
  sed -ne 's/^# \[\[\([^]|]*|\)*\([^]]*\)\]\].*/\2/p' | sort

Mám-li zase jméno obce, její souřadnice mohu získat tímto zaklínadlem:

m=Aš; curl -s 'http://maps.googleapis.com/maps/api/geocode/xml?address='"${m// /+},+CZ"'&sensor=false' |
  xmllint --xpath '//location[lat or lng]//text()' -

(Důležitý trik je to ,CZ, jinak bude Google znát spoustu Kolínů a Aš bude znamenat Americká Samoa. Alternativně si můžete z výsledků vyfiltrovat ty české pomocí XPath //result[address_component/short_name/text()="CZ"]/geometry/location[lat or lng]//text().)

Teď už to pro vygenerování jednoduchého CSV stačí spojit dohromady:

curl 'http://cs.wikipedia.org/w/index.php?title=Seznam_obc%C3%AD_s_roz%C5%A1%C3%AD%C5%99enou_p%C5%AFsobnost%C3%AD&action=edit' |
  sed -ne 's/^# \[\[\([^]|]*|\)*\([^]]*\)\]\].*/\2/p' | sort |
  while read m; do
    echo -n $m
    curl -s 'http://maps.googleapis.com/maps/api/geocode/xml?address='"${m// /+},+CZ"'&sensor=false' |
      xmllint --xpath '//location[lat or lng]//text()' - |
      tr -s '\n' ' ' | tr ' ' ','
    echo
    sleep 0.1
  done | sed 's/,$//'

Rádi byste hotové CSV?

Bonus: Podobně vygenerované CSV s pražskými částmi (katastrálními územími).

Bonus 2: A ještě CSV s obcemi s přenesenou působností (další velké obce a města)

Chido has two mouse-pets (Acomys Caihirinus, actually) and we finally did what we already planned to do long ago:

vlc http://pasky.or.cz:8090/mouses.flv

A live video stream of our mouse palace!

Some technical trivia – the IP cam used is Edimax IC-3110 (it’s pretty crappy, not recommended) and we are restreaming using vlc invocation:

cvlc -L --sout "#transcode{vcodec=mp4v,vb=1024,scale=1}:duplicate{dst=http{mux=ts,dst=:8090/mouses.flv},select=noaudio}" --no-sout-rtp-sap --no-sout-standard-sap --sout-ts-shaping=1000 --sout-ts-use-key-frames --ttl=40 rtsp://admin:PASSWORD@192.168.6.X:554/ipcam.sdp

(I did not get h264 FLV stream working reliably, unfortunately. I tried #transcode{vcodec=h264,venc=x264{keyint=20},vb=4096,scale=1} and duplicate{dst=http{mux=ffmpeg{mux=flv},dst=...,select=noaudio}.)