Taddeusz

Keep in mind also that printing support didn't get added until June this year at which point 6.5.2 was the latest release. I can't reliably say that it will work prior to 6.5.

I think you've lost me there. I don't remember what version of unRAID was available when I took over this Docker but it was never anything that old. Judging by my initial release date probably 6.3.2. Unfortunately, I have no way to test or really even to support on such an old version of unRAID. I don't know why there would be such a difference in the Docker subsystem but at this point that is the only major difference. Since Docker containers use the Linux kernel of the host system it is entirely possible that there was a vital kernel change that prevents Ghostscript from working correctly. I still have no way to test where that change occurred. <soapbox>I understand the attitude of "if it works don't fix it" but there have been some notable security fixes between then and now.</soapbox>

I'm running Chrome here at work. I enabled printing on my RDP connection. When I printed to the Guacamole "printer" it initiated a download of the resulting PDF. I even tried to change the option to always prompt for the download location and the dialog popped up in front like it should. What version of Chrome are you running? I'm running the latest, v69.

If you go to the terminal and run "top" what process is using so much CPU?

I can now verify that re-branding through an extension definitely works with this Docker.

Remote printing should be working. There was a fix some time ago that enabled that functionality. I've never personally used it so honestly I'm not sure how it is supposed to work. The latest version will give you that functionality. I will begin work on 1.0.0 when it reaches "release candidate" status. The way this is written it can only pull distinct releases that are available from apache.org. I'm not sure of their schedule but judging by the Github activity it should be "any day now". They have obviously been preparing for the 1.0.0 release. I've thought about doing the re-branding but I've never attempted it. Should definitely be possible though. Just drop the extension in your guacamole/extensions folder.

“%20” is the URL encoding for a space so check that you don’t have some trailing spaces on that line. For that matter you should probably check that you have no trailing spaces on any lines.

guacadmin for both username and password.

Thank you. I significantly bumped up the interval so it's not banging away at my cache drive so often. Not really necessary to be checking the drive every 5 seconds.

I've noticed when I leave my MakeMKV Docker running and the automatic disc ripper is enabled there seems to be a pretty constant amount of write activity on my cache drive. It seems like it's contributing about 6-8MB/s in write activity. Not much but I'm just trying to track down what applications are causing disk I/O. Seemed like an odd application to be causing so much write activity.

I've looked at the intel_pstate driver docs here: https://www.kernel.org/doc/html/v4.14/admin-guide/pm/intel_pstate.html I tried modifying the min_perf_pct in the /sys/devices/system/cpu/intel_pstate folder but that doesn't seem to do anything. If I understand the docs correctly here: This seems to indicate that the performance governor should run the CPU continuously at the maximum speed. Why isn't this happening in unRAID?

I'm trying to understand the CPU frequency scaling governors. My unRAID server runs on an i5 3470. Base frequency is 3.2Ghz, minimum CPU frequency is 1.6Ghz, and Turbo Boost can go up to 3.4Ghz quad or 3.6Ghz single. While Power Save does what I expect running the CPU most of the time at 1.6Ghz the Performance governor still scales the CPU frequency. It doesn't appear to get as low as the Power Save governor though. I'm used to the Windows "High Performance" power plan that seems to disable Intel's Speed Step and run continuously at the base frequency unless it needs more and kicks in the Turbo Boost. Is there any way to get the same kind of "High Performance" mode that Windows provides?

Yeah, I ended up installing xrdp on the Linux VM's just so I could remote into them without getting the disconnects. Actually finally ended up installing a Windows VM mainly because software I needed wasn't available on Linux.

From the dashboard screen you can acknowledge the warning by clicking on the warning icon.

I've been trying to figure out the performance issues I've been having with unRAID since I started using it. Running a cross-flashed IBM M1015 with 8 array drives, single parity, and a single drive cache. The parity and cache drives are attached to my motherboard's 6Gb connectors. I'm running the latest version of unRAID and used the modified 2.2 script. The problem I've been having is that during times of high disk usage like copying files between drives Docker application and VM performance drop significantly to the point where they become unresponsive until the disk operation is finished. Would it be normal to run the normal 2 minute test from 384 to 2944 without any discernible difference in resulting performance? With only a few low outliers it reported a pretty consistent 116.7-116.9 MB/s.

I use mostly Docker based applications and even those get preempted. I realize unRAID is primarily for storage and it kind of feels like it prioritizes array and cache over anything else. Like with Sonarr further downloads still happen while its unpacking the previous one but download performance is greatly diminished. So it’s not a complete freeze but it’s definitely favoring drive access over everything else.

I saw a topic talking about this with 4 cache drives. I only have a single cache drive. Even before I had no cache drive and a different CPU and motherboard I've always had this problem. I've tried adjusting the tunables. No matter what I do I still get these kind of "freezes". It often manifests itself when Sonarr is unpacking a download. Everything else seems to freeze until it's done unpacking the file. I can also trigger it by copying large files between array drives. This happened before and after I had a cache drive and also when I had an older dual Xeon system and now with my i5 3470. I have an IBM M1015 that's been cross flashed to IT mode. My cache drive and parity drive are on my motherboard's 2 6Gb connectors and the rest of the 8 are on the SAS controller. The only common factor besides the drives themselves has been the SAS card but I've tried swapping out for 2 2-port SATA cards and using all the 3Gb ports on the motherboard and it still happens. This doesn't happen that often but it is frustrating when it does happen.

Have you tried using the official Plex Media Server image? plexinc/pms-docker:latest

Looks like Apache is preparing Guacamole for a 1.0.0 release. Not sure when it will be released since there have not yet been any RC's. I'm a little excited about this next release as there are changes that I think will make the final image a bit smaller. They are definitely still aware of the VNC connection issue but it doesn't appear to be addressed yet.

I'm glad you figured it out. Personally I think the organization of the settings are a bit confusing. I think having the proxy parameters above the connection parameters is more confusing because a person is more likely to see the proxy hostname and assume that's for the remote computer. It's even happened to me on occasion even though I'm aware of it.

Thank you. I checked out the selection rules language. I think I have a solution already. I just need to test it when I get home.

Does the auto ripper follow the set language settings? So if I have it set to English it will only rip those tracks. Also, it would be nice to have an option on the auto ripper to only rip forced subtitles.

I’m using ASUSWRT Merlin so I’m not sure this will work for you, but this is what I use. https://gist.github.com/dd-han/09853f07efdf67f0f4af3f7531ac7abf

I wasn't aware of this. It appears the documentation is spread out but here's where it explains it better: https://guacamole.apache.org/doc/gug/administration.html#idm140500640336672. Basically it looks like in the connection list in settings you hit the "+" next to the connection on which you want to allow sharing and click "New Sharing Profile".

The RDP and TLS encryption types are now considered legacy as NLA was introduced with Windows Vista. If you're connecting to a Windows machine there's actually a checkbox in the RDP settings to only allow connections via NLA. I believe it's defaulted to checked as that is the most secure connection type.