[Libre-soc-dev] nlnet grant for funding maintenance/development of website/bugzilla

lkcl luke.leighton at gmail.com
Wed Nov 24 18:55:42 GMT 2021

On November 24, 2021 6:31:37 PM UTC, Reuben Lifshay <rlifshay at gmail.com> wrote:


hiya reuben, nice to hear from you.

>We are primarily a software development, web services, and managed
>hosting company. As such, project management is certainly something we
>are capable of and would be interested in, 

you understand what's needed, because you are Software Engineers: so are we.

> but as mentioned, that
>wasn't something we had discussed. My initial discussion with Jacob
>was in regards to the website (management of and/or
>replacement/improvement), hosting, infrastructure management, and
>DevOps software and management. 

yep, both Alain and I are perfectly capable of doing that, having each been managing servers for 20 years.

mythic-beasts hosting is sponsored (by mythic-beasts) and extremely low cost and high-end bandwidth for what it is (USD 15/month), it is definitely not worth trying to even spend time evaluating "improvements" because live-running migration would itself be extremely costly: again, for no good reason.

what is running is:

* nginx hosting fastcgid and static pages
* fail2ban in extreme draconian customised mode
* exim4 which is functional but annoying
* mailman2 which is again functional, customised to stop spammers
* postgresql for bugzilla
* bugzilla via fastcgid
* publicinbox which hooks into mailman2
* gitolite3 with ssh
* gitweb customised with colour syntax highlighing
* customised openvpn installation
* ikiwiki which autogenerates static web pages, 
   avoiding hacking attempts normally faced by
* dual off-site rsync backups
* DNS via bind9

all of this is "functional", is based on work i have been managing and running already for 15+ years, and is expected to stay that way.  loadavg remains around 0.4 on a dual core VPS.

about the only things that could be usefully considered would be chrooting of every service (isolation in case of attacks) and hitting exim4 with a hammer.

however even downtime of exim4 for even a few hours is extremely detrimental as it would bring down the mailing lists and publicinbox, as well as lose update notifications from bugzilla.

which is why i haven't touched it even though it is throwing tons of crap into undeliverable mode.

>We were discussing all of that
>primarily because of our interest in technology, but in doing so
>realized that RedShift might be a good fit, and would likely be able
>to improve things significantly. 

yeah, improvement is not necessary: anything at this point if "improved" would be a massive data-migration task and we could not realistically justify it given that the existing services are functional.

if they didn't, that would be an entirely different matter.

> We would definitely be interested in
>having a discussion with you to assess requirements and opportunities
>in more detail!

the server is functional, does its job, and is hard to justify asking NLnet for large funds to cover updating.  project management on the other hand is pretty urgent, takes up a significant amount of my time, and is easy to justify asking NLnet to help cover costs.


More information about the Libre-soc-dev mailing list