Automate Plugincheck Script running on the Server to detect newest versions.

NEW
Unassigned

Status

defect
--
major
6 years ago
5 years ago

People

(Reporter: rob1weld, Unassigned)

Tracking

Details

()

Cloned from Bug 850528.

Recently (see Bug 850212 and Bug 850528) we had a critical Update for Flash, and now one for Silverlight (on the same day) yet the PlugInCheck Page (https://www.mozilla.org/en-US/plugincheck/) said all was OK. 

Further comments in Bug 850212 Comment 5 claim that Versioning is inconsistent between different Browsers with the same Plugin.

We need to automate things in the Server's Script(s) so that some of the work is done for us thus reducing error and promoting timely fixes. 


We need "Version vs. OS Type" for Flash, and to consider that certain Plugins are only available on specific OSes (EG: Silverlight). 

We can also deal with all the "Unknown Plugins" (EG: "Windows Presentation Foundation", "Google Update", and "RealPlayer Download Plugin" are all unknown and have "Research Links").

Persons more knowledgable with the functioning of these mechanisms are welcomed to comment. I know little about what is on the Server and only see the result as a User.


Rather than checking every possible Website or burdening the Maintainers with an infinite List of Subscriptions we can check on all the Plugin versions Server-side.


All we need to do (easy to say - hehe) is have the Script running on the Server check if ANY User has a newer version of the Plugin installed than what the Script thinks is the newest version. When that occurs ONE mail can be sent to a small List of Maintainers, and they can manually check if our Script on the Server is OK or if it needs to be updated. A second mail could be sent the next day and/or the Maintainer could flag that Version for skipping (beta, not publicly available, data error, etc.).

This method is almost certain to work in every case. Only if no one obtains Updates, and subsequently does a PlugInCheck, could one of the Plugins slip by. It is certainly better than what we have now.


Thanks.
Depends on: 821279
Status: UNCONFIRMED → NEW
Ever confirmed: true
I have cited this bug several times:
e.g. on 2014-03-22, in bug 956905 comment # 128 and
on 2014-09-13, in bug 1020133 comment # 36.

Rob wrote:
> Rather than checking every possible Website or burdening the Maintainers with an
> infinite List of Subscriptions we can check on all the Plugin versions Server-side.
> 
> All we need to do ... ... is have the Script running on the Server check if ANY User
> has a newer version of the Plugin installed than what the Script
> [DJ-Leith: the 'plugincheck service'] thinks is the newest version. When that occurs
> ONE mail can be sent to a small List of Maintainers, and they can manually check if
> our Script on the Server is OK or if it needs to be updated.

I think perfidies.js can find "newer" versions.

It would be good if this concept;
an alert for 'new version of a known plugin has been detected', was implemented.


One obvious and well known issue is the release of a new version of a plugin.
This bug is to cover that frequent and important case.

There are other benefits.

Had this been implemented then situations like these next two bugs could ALSO
have been actioned sooner.

Bug 1084537 "Flash sometimes displayed as up to date whilst vulnerable, on Windows 7"
I think this has been caused by 'some infrastructure issue' - the 'Plugincheck Database',
in this case, HAS been updated BUT (as the screenshot shows) an 'old version' was
being reported as "Up to Date" - IN ERROR.

Had this been implemented then
many visitors to the 'plugincheck website', who had updated their Flash, would
be presenting a 'newer version' for checking.  This would have then triggered
ANOTHER 'newer version of plugin NAME has been seen' (Flash in this case)
BECAUSE the 'data sent from the Plugincheck Database' (in this case) was 'old data'.
So an investigation could be started as to WHY.

and
Bug 1083392 "Adobe Flash now 15.0.0.189 - plug in check database needs updating
[reported version is 15.0.0.152]"

DJ-Leith
You need to log in before you can comment on or make changes to this bug.