Skip to main content

scratch containers

I'm going to keep this real short. A scratch container is a base image that has nothing in it. It's effectively an empty file. Both Docker and rkt have scratch-like base images.

Here is a docker scratch file:

FROM scratch
ADD ca-certificates.crt /etc/ssl/certs/
ADD bin/HelloWorldOnce /HelloWorldOnce
CMD ["/HelloWorldOnce"]


Here is an acbuild script:

acbuild begin 
acbuild set-name myreg/helloworldonce
acbuild copy bin/HelloWorldOnce /bin/HelloWorldOnce
acbuild copy ca-certificates.crt /etc/ssl/certs/ca-certificates.crt
acbuild copy resolv.conf /etc/resolv.conf
acbuild set-exec /bin/HelloWorldOnce
#acbuild port add www tcp 5000
acbuild label add version 0.0.1
acbuild label add arch amd64
acbuild label add os linux
acbuild annotation add authors "Richard  <richard@......net>"
acbuild write --overwrite HelloWorldOnce-0.0.1-linux-amd64.aci
acbuild end


Why scratch? (see minimal containers and docker scratch official image) And for comparison read the bullshit justification for the Phusion baseimages.

I have long-since forgotten the official story but as memory serves the container developers that I know have endorsed statically linked applications running in a minimal container; like scratch. There are a few reasons that jump out at me. The first is fewer moving parts and so less heat. 2nd fewer dependencies if any. Finally; with fewer moving parts there are fewer attack vectors. For example if there is no ssh server in the container then one cannot attack the container through ssh.

Here are some notes that I discovered as I was trying to get parity between my docker and rkt containers:

  • the app in the container is a database client. It connects to a remote and public database using a published DNS address.
  • In order to get my application to connect to the DB using docker I needed to add the ca-certificates during the build.
  • In order to get my application to connect to the DB using rkt I needed to add /etc/resolv.conf to the container. 
I suppose I could have mounted the the host's resolv.conf volume in rkt but that might create different issues. By putting /etcd/resolv.conf in the container it cannot be spoofed at that level.

Comments

Popular posts from this blog

Entry level cost for CoreOS+Tectonic

CoreOS and Tectonic start their pricing at 10 servers. Managed CoreOS starts at $1000 per month for those first 10 servers and Tectonic is $5000 for the same 10 servers. Annualized that is $85K or at least one employee depending on your market. As a single employee company I'd rather hire the employee. Specially since I only have 3 servers.

The pricing is biased toward the largest servers with the largest capacities; my dual core 32GB i5 IntelNuc can never be mistaken for a 96-CPU dual or quad core DELL

If CoreOS does not figure out a different barrier of entry they are going to follow the Borland path to obscurity.

UPDATE 2017-10-30: With gratitude the CoreOS team has provided updated information on their pricing, however, I stand by my conclusion that the effective cost is lower when you deploy monster machines. The cost per node of my 1 CPU Intel NUC is the same as a 96 CPU server when you get beyond 10 nodes. I'll also reiterate that while my pricing notes are not currently…

Agile is still dead and has been since 1991

[updated 2011.09.30] yet another response to Agile is good.
When you have so much of you career invested in something like Agile, XP etc... it can be hard to see the forest for the trees. I had a consulting job in The Haag many years ago. IBM was the incumbent contractor at the customer site (a bank) but after 5 years on the job they had not written a single line of functioning code. In the office there were two teams of software people... both behind closed doors. The first team was the Data team and the second team was Functional. They rarely spoke and they never shared information. I was there for a week, introduced the client to OO and we had a functioning prototype. Smart people do smart things, You cannot make an underachiever exceptional by using Agile. Either they get "it" or they don't.
I just commented on a blog. I'm sure there is some validity to his post beyond observing that Agile Scrum is broken. It certainly is not what it was originally intended but for…

eGalax touch on default Ubuntu 14.04.2 LTS

I have not had success with the touch drivers as yet.  The touch works and evtest also seems to report events, however, I have noticed that the button click is not working and no matter what I do xinput refuses to configure the buttons correctly.  When I downgraded to ubuntu 10.04 LTS everything sort of worked... there must have been something in the kermel as 10.04 was in the 2.6 kernel and 4.04 is in the 3.x branch.

One thing ... all of the documentation pointed to the wrong website or one in Taiwanese. I was finally able to locate the drivers again: http://www.eeti.com.tw/drivers_Linux.html (it would have been nice if they provided the install instructions in text rather than PDF)
Please open the document "EETI_eGTouch_Programming_Guide" under the Guide directory, and follow the Guidline to install driver.
download the appropriate versionunzip the fileread the programming manual And from that I'm distilling to the following: execute the setup.sh answer all of the questio…