Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

My first deploy #1783

Closed
paulmelnikow opened this issue Jul 17, 2018 · 10 comments
Closed

My first deploy #1783

paulmelnikow opened this issue Jul 17, 2018 · 10 comments
Labels
operations Hosting, monitoring, and reliability for the production badge servers

Comments

@paulmelnikow
Copy link
Member

Hey maintainers,

I'd like to attempt a deploy on my own: see #1742 (comment) and #1742 (comment).

Would someone like to tag-team with me? The most helpful thing would be to monitor Sentry, though I know @espadrine hasn't been comfortable with anyone else having access to that. I think you could help in other ways, though, by hitting the server directly, monitoring the issues queue / twitter, and looking at the server status page.

This could be an evening this week, Oregon time, or Sunday afternoon, Chicago time. I've some availability next week too. Let me know if you're interested and let's see if we can find a time.

@paulmelnikow paulmelnikow added the operations Hosting, monitoring, and reliability for the production badge servers label Jul 17, 2018
@paulmelnikow
Copy link
Member Author

Just wanted to add: let's not cram to get stuff merged. If this deploy goes well let's plan on doing another one in a week.

@RedSparr0w
Copy link
Member

I'd recommend during the weekend if possible, as we have pretty good uptimes.

I could modify my script at http://shields.redsparr0w.com/1568/ to send an email/notification during the time period if it detects any notable downtime?

@chris48s
Copy link
Member

I should be available to keep an eye on things if you do Sunday afternoon central time. Evening pacific time doesn't co-ordinate well with my timezone.

Just in terms of what timezones maintainers are in (useful to know if we're trying to schedule stuff), based on the timestamp on everyone's last commit in the git log:

@chris48s +0100
@platan +0200
@PyvesB +0100
@RedSparr0w +1200

Might be easiest to co-ordinate on Discord for this one.

@paulmelnikow
Copy link
Member Author

Mine's -0400 when I'm home, though I'm often working on the road, between -0500 and -0700. It's interesting to see everyone's time zones!

@RedSparr0w Maybe you could send the notifications to a new discord channel?

@chris48s Cool! Let's plan on Sunday.

@PyvesB
Copy link
Member

PyvesB commented Jul 17, 2018

Haven't yet planned my week-end, but I'll probably be able to keep an eye on things at times. 😉

@RedSparr0w
Copy link
Member

RedSparr0w commented Jul 19, 2018

Can have the notifications like this:
(note the last 12 hours were up & down a lot more than usual)
chart 1
image

If the server has an uptime of <= 50% it will be marked as Unstable.
If the server has an uptime of >= 80% it will be marked as Stable.
Wait at least 10 minutes since the last notification before sending another one.
All the uptimes are based on request made within the last 5 minutes.

Possible changes to be made:

Setting Old New
Server Stable >= 80% >=90%
Server Unstable <= 50% <=40%
Min time between notifications 10 minutes 15 minutes
Group uptime by 5 minutes 5 minutes?

@espadrine
Copy link
Member

espadrine commented Jul 22, 2018

I'm eager to tag along! I am in CEST (+0200).

I planned on doing an update (had a bit of computer trouble at home this month). I can do an update today if we fail to schedule something up.

@paulmelnikow
Copy link
Member Author

Cool! Things are lined up for when @chris48 is gets online, which is a little over an hour from now, shortly after 10 a.m. UTC-5. Will be great for you to be around in case anything goes dreadfully wrong! At Chris's suggestion, let's coordinate on Discord.

@paulmelnikow
Copy link
Member Author

s0 is live. We're going to monitor and if all looks good in a few hours we'll deploy to the other servers and the frontend.

@paulmelnikow
Copy link
Member Author

And… it's live!

screen shot 2018-07-22 at 4 39 28 pm

Thanks & congrats, everyone!

If anything systemic comes up, feel free to post here and one of the maintainers can reopen.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
operations Hosting, monitoring, and reliability for the production badge servers
Projects
None yet
Development

No branches or pull requests

5 participants