Taddeusz

I apologize for the problems. This time it's really fixed. Tomcat was really not starting when the /config/guacamole/log folder didn't exist. I had to change the point at which that folder was being created.

When I say exactly, I mean does the page load but it's just white with no errors. Or do you get a 404 or other error trying to load the page? What permissions are on your ApacheGuacamole appdata folder?

Exactly what is it doing?

For those having problems let me know what the exact problem is. Does the login form show or do you get a white page and no visible error? Do any errors show in your Docker log or catalina.out? I believe there are problems with file permissions going on but I guess I'm not quite understanding the root of the problem.

I've pushed out a change to rollback the permission updates. The redirected catalina.out is still there.

Well, had my newest drive offlined Friday night. Spent a lot of yesterday troubleshooting. I've come to the conclusion that there's something wrong with the backplane in my drive cage. It's a Rosewill RSV-SATA-Cage-34. I removed the backplane and rigged up some fans for cooling. Direct connected all the drives. It ran a rebuild just fine with no more UDMA CRC errors. I tried to replace the thermal compound on my SAS card but they used thermal glue so it's not coming off. I've got a couple drive cages coming but they don't have backplanes. Also Rosewill but really low cost.

The fail message after starting tomcat has always been there. I’m not sure why it shows that. Does the page load but just shows nothing? If so, that means it can’t connect to your authentication source. Look at the catalina.out file and see if there are any errors there.

I have updated the the image to output Tomcat's catalina.out file /config/guacamole/log to help troubleshooting. I also changed the scripts to always update file permissions to try to mitigate a problem where it appears that the database can't be read. Please let me know if you're having any issues.

I know that card gets hot. It just has a heatsink on it. I'll look at putting a fan on. I'll at least put some fresh thermal compound on there. It could probably use it.

I guess I have some digging to do this weekend.

I'm on 6.5.2 and a week ago I started getting a lot of these errors and then write errors which offlined my newest drive, only in service a month. I replaced the drive and have been fine until now I received 1 UDMA CRC error with the brand new drive. Diagnostics on the "old" new drive passed on a different system with only 3Gb/s SATA. In fact, I just RMA'd the drive and shipped it today. I had actually tried to put it back in service but as soon as unRAID booted back up it started to get UDMA CRC errors again on a different tray. I'm running an IBM M1015 reflashed to an LSI 9211-8i. First time I've ever seen this kind of thing. Since this is something that all of a sudden just started happening I'm not sure what to do. The two drives I've been having issues with are in a 4 slot drive cage. The other 5 drives are direct connected. The drives I've been having trouble with are new old stock HGST 7K4000 2TB drives. Not sure whether the cage or the controller or SAS to 4xSATA cable could be the issue? I have a shorter cable I could try. For now it's just a single error so nothing has gone offline again. This is still troubling.

Ok, give me a few moments and I'll publish it to latest. Then you can change your tag back and update.

Ok, change your docker to "jasonbean/guacamole:0.9.14gs". Let me know if that changed things.

Ok, if you don't mind briefly running a test version I can push out one with gs included later today.

I'm trying to run High Sierra from a VM without pass-through GPU but get a lot of video corruption. Especially from Google Chrome. Any way to solve this?

Just to give some closure on my own issue I have moved on to Nextcloud. It is a fork of ownCloud and, IMHO, is better. With Nextcloud everything works as expected. It also has the benefit of using local storage as "external". It just all works.

Ok, I guess the problem I'm actually having is that I have a "Data" share on my unRAID server that I am mounting in ownCloud via SMB. When I try to save files to this SMB share it fails.

I'm having trouble running this through my let's encrypt nginx server. I'm running the app on my iPad. I can login and view files but get errors when it tries to upload files. I tried following these instructions which got me to this point: https://doc.owncloud.org/server/9.0/admin_manual/configuration_server/reverse_proxy_configuration.html My nginx configuration is as follows: location ^~ /owncloud/ { include /config/nginx/proxy.conf; proxy_pass https://192.168.22.90:8443/; proxy_buffering off; } My ownCloud config has these values: 'trusted_proxies' => ['192.168.22.90'], 'overwritehost' => '<my public hostname>', 'overwriteprotocol' => 'https', 'overwritewebroot' => 'owncloud', 'overwritecondaddr' => '^192\.168\.22\.90$', What am I doing wrong?

In mine the map directive is at the very top of my default file outside the server directives.

Since you've moved to Nginx here's my reverse proxy settings for my own: map $http_upgrade $connection_upgrade { default upgrade; '' close; } location ^~ /guacamole/ { proxy_pass http://<replace your ip>:<replace your port>/guacamole/; proxy_buffering off; proxy_http_version 1.1; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header Upgrade $http_upgrade; proxy_set_header Connection $http_connection; access_log off; } The proxy_buffering off is very important if you read their documentation.

No, this Docker does not contain the Apache HTTP Server. You would need to install the separate Apache Server Docker to do reverse proxy and other web hosting duties. The only thing this includes is Apache Tomcat to run the Guacamole Client. Tomcat is built to run Java servlets. It's not designed to fill the duties of the Apache HTTP Server.

Doesn't matter. If you're looking at your CPU usage on the unRAID Dashboard the overall load is going to be averaged over all 8 threads. Try looking at top, the vm process for my Windows 10 VM shows a nearly constant 33% usage. Top is different. It calculates load based on a single CPU. So a single process on a quad core CPU can have up to 400% load.

It may be because you have 8 threads. Your overall idle usage would be less on your 8 thread Xeon than on my 4 core 4 thread i5 because it would be averaged over 8 rather than 4 CPU's.

I like the new Docker console option but it would be nice to make the shell it launches selectable. Most, if not all, Docker images I've had to go into have bash installed. Bash is much preferred to the bare "sh".

Not sure about ddclient but I figured out how to add a script to my router to do the same thing, and it uses their v4 API.