[
https://issues.apache.org/jira/browse/GUACAMOLE-1537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17491141#comment-17491141
]
Mike Jumper commented on GUACAMOLE-1537:
----------------------------------------
Yes, the 404 status is used by the REST API to indicate that a particular
resource does not exist. It is common within the webapp to attempt to retrieve
a resource that _might_ exist, and then handle the fact that it doesn't
appropriately.
You definitely should not assume that error codes from the REST API indicate
errors in the application, nor that it's safe to intercept and rewrite errors
in general. Guacamole expects to be able to read the machine-readable content
of its own responses.
The only case where an error code indicates an error in the application are the
5xx family of codes, and even then you definitely should not attempt to
intercept and modify these.
> Guacamole drops error 403 issue with traefik errorhandler
> ---------------------------------------------------------
>
> Key: GUACAMOLE-1537
> URL: https://issues.apache.org/jira/browse/GUACAMOLE-1537
> Project: Guacamole
> Issue Type: Bug
> Components: guacamole
> Affects Versions: 1.4.0
> Reporter: averon
> Priority: Critical
> Original Estimate: 1h
> Remaining Estimate: 1h
>
> Hello there,
> I am using guacamole with reverse proxy traefik. I have a middleware which
> catches the http errorcodes. Somehow the guacamole drops an error 403 when
> you enter the login page. That results in a critial error, where you can not
> access the guacamole login page anymore.
> I have a workarround to not catch error 403. Is this maybe possible to fix?
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
