[
https://issues.apache.org/jira/browse/YARN-4767?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Daniel Templeton updated YARN-4767:
-----------------------------------
Attachment: YARN-4767.002.patch
This patch also works for single-node clusters. The basic idea is that the
{{AmIpFilter}} appends /redirect to the URL when it redirects to the web proxy.
(It only does that when it's contacted by a host other than the web proxy.)
The web proxy looks for the /redirect when it receives a request. If it's
there, and if the request originated from the web proxy itself, it instead
redirects to an error page, ending the endless loop.
The error page is implemented within the app page and is triggered by appending
an "R" to the end of the app ID. If the app page sees the "R", it prints the
error instead of the app page. This part is the part I'm least confident about
being a reasonable thing to do. Feedback welcome.
> Network issues can cause persistent RM UI outage
> ------------------------------------------------
>
> Key: YARN-4767
> URL: https://issues.apache.org/jira/browse/YARN-4767
> Project: Hadoop YARN
> Issue Type: Bug
> Components: webapp
> Affects Versions: 2.9.0
> Reporter: Daniel Templeton
> Assignee: Daniel Templeton
> Priority: Critical
> Attachments: YARN-4767.001.patch, YARN-4767.002.patch
>
>
> If a network issue causes an AM web app to resolve the RM proxy's address to
> something other than what's listed in the allowed proxies list, the
> AmIpFilter will 302 redirect the RM proxy's request back to the RM proxy.
> The RM proxy will then consume all available handler threads connecting to
> itself over and over, resulting in an outage of the web UI.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
