init scripts for carbon-c-relay

[eholzbach]

Would be helpful if there was an option to run this process as a daemon, allowing easy init control.

[grobian]

Interesting one. It's not too hard, but since tools like systemd apparently don't want this to happen, I figured I didn't have to do anything here. Where I work, we just have an init script that redirects its output to a log and backgrounds the process, or as alternative setup we use things like https://pypi.python.org/pypi/supervisor.

[eholzbach]

ehh, yeah. ill just supervisord it. have it everywhere anyway.

[mhollick]

Hi, I use daemonize. I have put together an init file and rpm spec file. I will do a pull request over the weekend if you think they could be handy.

Mat. On 5 Sep 2014 19:06, "eric" notifications@github.com wrote:

Closed #13 https://github.com/grobian/carbon-c-relay/issues/13.

— Reply to this email directly or view it on GitHub https://github.com/grobian/carbon-c-relay/issues/13#event-161884461.

[grobian]

Might be a good thing to share. Please do, I'll compare it to mine, see if we can improve both.

[pkittenis]

Hi,

Have a chkconfig compatible init script that uses start-stop-daemon. daemonize is another option but that is redhat specific.

At https://gist.github.com/pkittenis/9bd2144afbed4f8b9e17

[davewongillies]

I've got debian packaging too which I'm happy to contribute (been meaning too). It uses start-stop-daemon for its initscript.

[mhollick]

I have created a pull request with an init file and some other bits and bobs. davewongillies and pkittenis have a good point about daemonize being a RedHat-ism. I did not realise this.

[grobian]

Ok, this feels like we need to agree on how the binary is called. Now I admit I made a mess of it myself already, but what I see now is:

Makefile: relay our rpms: carbon-c-relay pkittenis: carbon-c-relayd(script) -> relay mhollick: cc_relay davewongillies:

[davewongillies]

I've built mine at a previous job as ccrelayd but I'm happy with carbon-c-relay. Actually, almost anything besides just relay since its so ambiguous. Perhaps some more wordplay on the whole carbon thing and removing the language that its written in from its name. Apparently C2 (carbon-c, geddit? :smile: ) is diatomic carbon, how about diatomicd or diatomic-relay?

[mhollick]

• I thought that I had removed all evidence of the cc_relay name, not really happy with it. I concur that 'relay' is ambiguous.

As an experienced bike shedder; I would like to suggest 'carbon-router'. And it should be painted green.

[grobian]

hah, wow :) I think relay is its main purpose, but since it does more these days, perhaps good thing to drop from its name?

[grobian]

Ok, following dave's lead here: (courtacy wikipedia)

carbene carbyne (3C groups) chaoite (white carbon, which' existence is disputed) cubane (8 carbon atoms in a cube) anthracite (highest grade of coal after graphite) calcite (transparent, shows things double when viewed through it)

Fromt his list, I like cubane because it would refer somewhat to one of the original objectives of carbon-c-relay to be multithreaded by default (iso needing multiple procs with a lb in front). I also like what wikipedia mentions more: "Cubane also has the highest density of any hydrocarbon, further contributing to its ability to store large amounts of energy, which would reduce the size and weight of fuel tanks in aircraft and especially rocket boosters." :)

[filippog]

out of curiosity has this been decided upon? looks at bikeshed I'm asking because I'm planning to upload it to debian and transitioning later while not impossible is just confusing to users

[grobian]

I don't think changing the name of the project does any good, so I'll leave it at carbon-c-relay. I've merged the init script branch by mhollick, so I consider this issue as closed.

[filippog]

:+1: carbon-c-relay works just fine IMO