hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Elek, Marton (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-12961) Ozone: Use relative resource urls in KSM/SCM web ui
Date Sat, 23 Dec 2017 13:46:06 GMT

    [ https://issues.apache.org/jira/browse/HDFS-12961?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16302438#comment-16302438

Elek, Marton commented on HDFS-12961:

Patch is uploaded. To test:

1. First of all check ksm/scm ui in the usual way. It works as before.

2. Try it out with a reverse proxy:

I tested with caddy (single binary webserver) with the following config:

http://localhost:1234 {
  tls off 
  proxy /scm http://localhost:9876 {
     without /scm
  proxy /ksm http://localhost:9874 {
     without /ksm

And starting the server:

./caddy -conf Caddyfile

Webui worked with http:/localhost:1234/scm and http:/localhost:1234/ksm (and also with the
normal way: http://localhost:9874)

> Ozone: Use relative resource urls in KSM/SCM web ui
> ---------------------------------------------------
>                 Key: HDFS-12961
>                 URL: https://issues.apache.org/jira/browse/HDFS-12961
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: HDFS-7240
>            Reporter: Elek, Marton
>            Assignee: Elek, Marton
>         Attachments: HDFS-12961-HDFS-7240.001.patch
> The current SCM/KSM web ui (+namenode/datanode) uses absolute urls:
> Css is defined as /static/..., http calls use /jmx as url, etc.
> This is problem with typical cloud native environments where some components could be
behind a reverse proxy. 
> For example in kubernetes the component could be accessible via the kubctl proxy and
specific url:
> http://localhost:8001/api/v1/namespaces/default/services/ozonetest-http-ksm:web/proxy/
> This url points to the scm web ui but all of the resources are loaded from http://localhost:8001
as we use absolute urls.
> I suggest to use relative url everywhere. It could work both with standalone cluster
(were the default context is the /) and with proxied environment (where the scm location could
be different).

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org

View raw message