[jira] [Commented] (GUACAMOLE-1271) Debugging failing guacamole client container in AWS ECS

["Mike Jumper (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/GUACAMOLE-1271?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17270756#comment-17270756 ] 

Mike Jumper commented on GUACAMOLE-1271:
----------------------------------------

I don't believe there is anything wrong with the latest version of the {{guacamole/guacamole}} image:

* Deploying the latest image myself, things start up correctly.
* It was tested prior to release and definitely, at a minimum, starts up. The regression tests performed against the image included logging in using the various supported databases, managing connections, establishing connections using the various supported protocols, etc.
* If there is a problem with the image that made it past testing and into release, I would expect it to be more subtle than failing to start.

Can you provide the environment variables that you're setting for the container?

{quote}
With recent attempt to switch to version 1.3.0 the client containers seem to die upon start with the following error:
{quote}

In the log excerpt provided, there is no error. The sequence of events is:

# The webapp is deployed ("Deployment of web application archive [/usr/local/tomcat/webapps/guacamole.war] has finished in [4,873] ms")
# Tomcat finishes startup ("Server startup in 4981 ms")
# ... two and a half minutes pass ...
# Tomcat begins orderly shutdown ("Stopping service [Catalina]")
# Tomcat warns that there is a thread still running ("The web application [guacamole] appears to have started a thread named [com.google.inject.internal.util.$Finalizer] ...") - this is a known issue with the version of Google Guice used (dependency injection library) and can be safely ignored. It doesn't cause the webapp to stop, it's a warning that occurs when the webapp is _already stopping_.
# Shutdown completes

{quote}
I'm a bit at a loss as to how to debug this further. Guacd has an option for additional debugging that can be enabled via an environment variable, but I can't seem to find one for the client.
{quote}

http://guacamole.apache.org/doc/gug/configuring-guacamole.html#webapp-logging

For the Docker image, the documentation for the variable seems to be missing from the manual, but it's {{LOGBACK_LEVEL}}. If 

> Debugging failing guacamole client container in AWS ECS
> -------------------------------------------------------
>
>                 Key: GUACAMOLE-1271
>                 URL: https://issues.apache.org/jira/browse/GUACAMOLE-1271
>             Project: Guacamole
>          Issue Type: Bug
>          Components: guacamole-client
>    Affects Versions: 1.3.0
>         Environment: Amazon ECS (Amazon Linux 2)
>            Reporter: Dominik Borkowski
>            Priority: Minor
>
> Hello,
> We use both the client and server guacamole images in Amazon ECS. With recent attempt to switch to version 1.3.0 the client containers seem to die upon start with the following error:
> {code:none}
> 19-Jan-2021 20:17:21.304 INFO [localhost-startStop-1] org.apache.catalina.startup.HostConfig.deployWAR Deployment of web application archive [/usr/local/tomcat/webapps/guacamole.war] has finished in [4,873] ms
> 19-Jan-2021 20:17:21.307 INFO [main] org.apache.coyote.AbstractProtocol.start Starting ProtocolHandler ["http-nio-8080"]
> 19-Jan-2021 20:17:21.316 INFO [main] org.apache.catalina.startup.Catalina.start Server startup in 4981 ms
> 19-Jan-2021 20:19:55.238 INFO [Thread-4] org.apache.coyote.AbstractProtocol.pause Pausing ProtocolHandler ["http-nio-8080"]
> 19-Jan-2021 20:19:55.254 INFO [Thread-4] org.apache.catalina.core.StandardService.stopInternal Stopping service [Catalina]
> 19-Jan-2021 20:19:55.285 WARNING [localhost-startStop-2] org.apache.catalina.loader.WebappClassLoaderBase.clearReferencesThreads The web application [guacamole] appears to have started a thread named [com.google.inject.internal.util.$Finalizer] but has failed to stop it. This is very likely to create a memory leak. Stack trace of thread:
>  java.lang.Object.wait(Native Method)
>  java.lang.ref.ReferenceQueue.remove(ReferenceQueue.java:144)
>  java.lang.ref.ReferenceQueue.remove(ReferenceQueue.java:165)
>  com.google.inject.internal.util.$Finalizer.run(Finalizer.java:114)
> 19-Jan-2021 20:19:55.296 INFO [Thread-4] org.apache.coyote.AbstractProtocol.stop Stopping ProtocolHandler ["http-nio-8080"]
> 19-Jan-2021 20:19:55.299 INFO [Thread-4] org.apache.coyote.AbstractProtocol.destroy Destroying ProtocolHandler ["http-nio-8080"]
> {code}
> Additional information:
>  * Affected image is `guacamole/guacamole:1.3.0`, while `guacamole/guacamole:1.2.0` works just fine, as do all the earlier versions
>  * Switching back and forth reliably shows that 1.3.0 fails, while 1.2.0 works fine
>  * `guacamole/guacd:1.3.0` works fine
>  * Local execution of `guacamole/guacamole:1.3.0` works fine
>  * Building from source code at branch 1.3.0 and using that with ECS shows the same exact error
>  
> I'm a bit at a loss as to how to debug this further. Guacd has an option for additional debugging that can be enabled via an environment variable, but I can't seem to find one for the client. 
> I'd appreciate any input, thank you in advance!



--
This message was sent by Atlassian Jira
(v8.3.4#803005)