Opened on 09/26/2017 at 10:19:12 AM
Closed on 10/20/2017 at 10:01:24 AM
Last modified on 10/27/2017 at 01:12:14 PM
#5788 closed defect (fixed)
Retest Notification AB machinery
Reported by: | wspee | Assignee: | |
---|---|---|---|
Priority: | Unknown | Milestone: | |
Module: | Unknown | Keywords: | |
Cc: | Blocked By: | ||
Blocking: | Platform: | Unknown / Cross platform | |
Ready: | no | Confidential: | no |
Tester: | Unknown | Verified working: | no |
Review URL(s): |
https://codereview.adblockplus.org/29556615/ |
Description (last modified by wspee)
Background
This is an issue to track the re-test of #2981.
What to change
Deploy the following notification:
From: https://codereview.adblockplus.org/29556615/
Notification: 3
severity = information target = extensionVersion>=10 extensionVersion<=1 [1] title.en-US = Test #2981 variant #1 message.en-US = This is a test notification to test #2981 variant #1 sample = 0.005 [2] title.en-US = Test #2981 variant #2 message.en-US = This is a test notification to test #2981 variant #2 sample = 0.005 [2] title.en-US = Test #2981 variant #3 message.en-US = This is a test notification to test #2981 variant #3 sample = 0.005
This notification should yield 3 variants that are downloaded by 0.5% of the clients but never be shown. The remaining clients should get an empty notification.json.
And see if the effects described in #2981 still occur.
Note
The test was started on Tue, 17 Oct 2017 16:50:22 +0200 and ended on Thu, 19 Oct 2017 13:58:09 +0200.
Data will be logged to access_log_notification on all filter servers.
Example excerpt of /var/log/nginx/access_log_notification
10.8.0.1 - - [17/Oct/2017:09:31:57 +0000] "GET /notification.json?lastVersion=201710161200-2%2F1 HTTP/1.1" 200 323 "-" "Wget/1.18 (linux-gnu)" "-" https "-" "notification.adblockplus.org" "-" 201710170931-3/0
Note how the new version string showing the group the client ended up in is appended to the end of the line.
Attachments (0)
Change History (12)
comment:1 Changed on 09/26/2017 at 10:22:07 AM by wspee
comment:2 Changed on 09/26/2017 at 10:44:07 AM by wspee
comment:7 Changed on 10/17/2017 at 02:50:42 PM by abpbot
comment:8 Changed on 10/17/2017 at 03:23:28 PM by wspee
The test started at ~Tue, 17 Oct 2017 16:50:22 +0200
comment:10 Changed on 10/19/2017 at 02:52:00 PM by abpbot
A commit referencing this issue has landed:
Issue 5788 - Disable notification to retest the notification AB machinery
comment:11 Changed on 10/20/2017 at 10:01:24 AM by wspee
- Description modified (diff)
- Resolution set to fixed
- Status changed from reviewing to closed
comment:12 Changed on 10/27/2017 at 01:12:14 PM by Kirill
I confirm that the data looks good. The distribution was ~0.0065 for all groups.
A commit referencing this issue has landed:
Issue 5788 - Added notification to retest the notification AB machinery