I was wondering if anybody would have the time to create some sort of application status dashboard, similar to the ones found on google (http://www.google.com/appsstatus#hl=en) or amazon http://status.aws.amazon.com/). Essentially, something that acts like a simplified external facing blog, where people could update the different pieces as problems are detected. Eventually, it would be nice to be able to tie it with our different monitoring softwares, but to begin with, it would be very convenient for our partners to be able to see the overall status of our different layers. For instance, if someone detects that m.wikipedia.org is not working, the first step would be to update said dashboard to inform the less technically-savy people who are not necessarily on IRC of the problem and that someone is looking at it / in the process of fixing it. Another important feature would be to keep history on problems that have happened in the past, much like google does. I know this is already done to some extent with the server admin log, but having an easy to read interface would in my opinion prove beneficial.
Anyway, any suggestion on additional features, or requirements are welcomed.
--Fred.
On Wed, Jan 13, 2010 at 2:57 PM, Fred Vassard fvassard@wikimedia.org wrote:
I was wondering if anybody would have the time to create some sort of application status dashboard, similar to the ones found on google (http://www.google.com/appsstatus#hl=en) or amazon http://status.aws.amazon.com/).
The toolserver also has something similar: http://status.toolserver.org/
Anyway, any suggestion on additional features, or requirements are welcomed.
I don't have any suggestions, but I just want to say that this is a good idea. :-)
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 37-01--10 03:59 PM, Fred Vassard wrote:
I was wondering if anybody would have the time to create some sort of application status dashboard, similar to the ones found on google (http://www.google.com/appsstatus#hl=en) or amazon http://status.aws.amazon.com/). Essentially, something that acts like a simplified external facing blog, where people could update the different pieces as problems are detected. Eventually, it would be nice to be able to tie it with our different monitoring softwares, but to begin with, it would be very convenient for our partners to be able to see the overall status of our different layers. For instance, if someone detects that m.wikipedia.org is not working, the first step would be to update said dashboard to inform the less technically-savy people who are not necessarily on IRC of the problem and that someone is looking at it / in the process of fixing it. Another important feature would be to keep history on problems that have happened in the past, much like google does. I know this is already done to some extent with the server admin log, but having an easy to read interface would in my opinion prove beneficial.
Anyway, any suggestion on additional features, or requirements are welcomed.
--Fred.
There is also http://ezyang.com/wikistatus/ which provides a pretty good model.
- -Mike
Fred Vassard wrote:
I was wondering if anybody would have the time to create some sort of application status dashboard, similar to the ones found on google (http://www.google.com/appsstatus#hl=en) or amazon http://status.aws.amazon.com/). Essentially, something that acts like a simplified external facing blog, where people could update the different pieces as problems are detected. Eventually, it would be nice to be able to tie it with our different monitoring softwares, but to begin with, it would be very convenient for our partners to be able to see the overall status of our different layers. For instance, if someone detects that m.wikipedia.org is not working, the first step would be to update said dashboard to inform the less technically-savy people who are not necessarily on IRC of the problem and that someone is looking at it / in the process of fixing it. Another important feature would be to keep history on problems that have happened in the past, much like google does. I know this is already done to some extent with the server admin log, but having an easy to read interface would in my opinion prove beneficial.
Anyway, any suggestion on additional features, or requirements are welcomed.
--Fred.
Requirements: *When the site goes down, the dashboard must still be up, no matter which layer failed. *The dashboard must be able to handle the slashdot effect that the site going down would have on it.
The dashboard would fix bug 20079, forked from 16043.
https://bugzilla.wikimedia.org/show_bug.cgi?id=20079 https://bugzilla.wikimedia.org/show_bug.cgi?id=16043
wikitech-l@lists.wikimedia.org