Package building for ARM
Finally, we’ve built a package for arm32v7 from recent Kotori sources, which should be suitable to be installed on your RaspberryPi machine running Raspbian Stretch. We used Docker for building this and the outcome looks promising already:
root@testbox-arm32v7:/$ cat /etc/debian_version 9.7 root@testbox-arm32v7:/$ uname -a Linux 5a95c418e610 4.9.87-linuxkit-aufs #1 SMP Wed Mar 14 15:12:16 UTC 2018 armv7l GNU/Linux
root@testbox-arm32v7:/$ /opt/kotori/bin/kotori --version Kotori version 0.21.1 root@testbox-arm32v7:/$ /opt/kotori/bin/kotori --config /etc/kotori/kotori.ini 2019-03-03T01:36:42+0000 [kotori ] INFO: Starting Kotori version 0.21.1 2019-03-03T01:36:42+0000 [kotori ] INFO: Using configuration file /etc/kotori/kotori.ini 2019-03-03T01:36:42+0000 [kotori.core] INFO: Enabling applications  2019-03-03T01:36:42+0000 [kotori.core] INFO: Enabling vendors 
The new .deb package for Kotori on ARM is available for manual download from . It is made on a Debian “stretch” 9.7 system, so it should probably also work best when installed to the same distribution.
Installing this would work like
wget https://packages.hiveeyes.org/hiveeyes/foss/testing/kotori-standard_0.21.1-1_armhf.deb dpkg -i kotori-standard_0.21.1-1_armhf.deb
If your system tells you about a conflict with the (previously installed) canonical
"kotori" package, you should be safe to just remove this by running
"apt-get remove kotori" first. The configuration files at
/etc/kotori should be left untouched, but you might want to do a simple backup of them before to account for “better safe than sorry” ;].
About the “
kotori-standard” flavor aka. downsizing the package
Please note while earlier packages of Kotori for ARM were about 42 MB in size, the new
"kotori-standard" package for ARM is a bit more lightweight clocking in with about 29 MB in download size. It will contain the dependency packages listed in both of the daq and export sections defined in Kotori’s
setup.py. The functionality offered by this scope should be sufficient for normal operations on a RaspberryPi and such. Please let us know when it lacks important features, then we will reconsider this approach.
Further testing required
As we used Docker for building this artifact, we haven’t been able to test it for real, so may we humbly ask if you could test this package on your RaspberryPi or a similar ARM-based SoC machine? Thanks in advance already, we are looking forward to your feedback.
With kind regards,