infra-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Michael Noll (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (INFRA-16771) Maven Central June states is incomplete
Date Mon, 16 Jul 2018 17:44:00 GMT

    [ https://issues.apache.org/jira/browse/INFRA-16771?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16545536#comment-16545536
] 

Michael Noll edited comment on INFRA-16771 at 7/16/18 5:43 PM:
---------------------------------------------------------------

To reproduce:

1. Go to https://repository.apache.org/#central-stat
2. Select "org.apache.kafka" as the "Project".
3. Set "Type" to "Unique IPs".
4. Set "Group ID" to "org.apache.kafka" (which should be pre-selected)
5. Set "Artifact ID" to "kafka-streams".  (see below: you can also select "kafka-clients",
which has the same issue)
6. Don't change the "Version" field (keep it at "Select a version...").
7. Click the "Export CSV" button right next to the "Version field".

You will then notice that the .csv file will show `0` for June, and the generated graph right
underneath the Group/Artifact/Version boxes "Unique IPs Over the Last 12 Months For org.apache.kafka"
will show `0` for June.  Same problem for the Artifact ID "kafka-clients" (see step 5).

Note: The stats website WILL show information such as Unique IPs per version in one of the
tables on the same screen, but these are "Unique IPs per version", and you can't sum the numbers
to get the "Unique IPs across all versions" (which is currently reported as `0`) because you
can't sum uniques.  That is, the sum of "Unique IPs per version" is very, very likely much
higher than the true number of "Unique IPs across all versions".


was (Author: miguno):
To reproduce:

1. Go to https://repository.apache.org/#central-stat
2. Select "org.apache.kafka" as the "Project".
3. Set "Type" to "Unique IPs".
4. Set "Group ID" to "org.apache.kafka" (which should be pre-selected)
5. Set "Artifact ID" to "kafka-streams".  (see below: you can also select "kafka-clients",
which has the same issue)
6. Don't change the "Version" field (keep it at "Select a version...").
7. Click the "Export CSV" button right next to the "Version field".

You will then notice that the .csv file will show `0` for June, and the generated graph right
underneath the Group/Artifact/Version boxes "Unique IPs Over the Last 12 Months For org.apache.kafka"
will show `0` for June.  Same problem for the Artifact ID "kafka-clients" (see step 5).

Note: The report WILL show information such as Unique IPs per version in one of the tables
on the same screen, but these are "Unique IPs per version", and you can't sum the numbers
to get the "Unique IPs across all versions" (which is currently reported as `0`) because you
can't sum uniques.  That is, the sum of "Unique IPs per version" is very, very likely much
higher than the true number of "Unique IPs across all versions".

> Maven Central June states is incomplete
> ---------------------------------------
>
>                 Key: INFRA-16771
>                 URL: https://issues.apache.org/jira/browse/INFRA-16771
>             Project: Infrastructure
>          Issue Type: Bug
>          Components: maven
>            Reporter: Guozhang Wang
>            Assignee: Chris Lambertus
>            Priority: Major
>
> The maven central June stats for Kafka is incomplete:
> https://repository.apache.org/#central-stat
> For example, kafka-clients and kafka-streams numbers for June are still 0.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message