Page MenuHomePhabricator

When installing instances, starting puppet runs etc.: ping the developer by mail about the status
Closed, ResolvedPublic

Description

Implementation idea:

send mails about the progress, and when certain steps have finished, and when finally finished.

The mail address is known, since the developer is known.


Version: unspecified
Severity: enhancement
See Also:
https://bugzilla.wikimedia.org/show_bug.cgi?id=44180

Details

Reference
bz45214

Event Timeline

bzimport raised the priority of this task from to Low.Nov 22 2014, 1:13 AM
bzimport added a project: Cloud-VPS.
bzimport set Reference to bz45214.

It already sends a mail when the instance creation has finished.

(In reply to comment #1)

It already sends a mail when the instance creation has finished.

I did *not* receive any mail about the progress of the instance generation which I started as a logged-in labsconsole user with confirmed e-mail address.

"Your e-mail address was authenticated on 21 February 2013 at 01:05."

Interesting, this is what I received a few days back:
Subject: Your instance is ready to be logged into.
The following instance has been created, and is ready to be logged into: webtools-login

(In reply to comment #3)

Interesting, this is what I received a few days back:
Subject: Your instance is ready to be logged into.
The following instance has been created, and is ready to be logged into:
webtools-login

Yes, interesting. Perhaps a glitch in the system, if you think, it's in, but did not work in my case, i will modifiy the title.

We have something in puppet that's supposed to send a mail, but it often doesn't work because a puppet run isn't 100% successful on the first run. However, we now have echo support for create, reboot and delete actions. For the create and reboot actions we need to add reactor support to our saltstack configuration.

(In reply to comment #5)

We have something in puppet that's supposed to send a mail, but it often
doesn't work because ...

This is what soem reported to me. So I filed this bug without (then) knowing that it was already implemented, but perhaps - from my perspective - failed. That's why I filed it. Can be now used for tracking problems around this feature.

We're now doing this with echo notifications and it is working reliably.