lukeoslavia
-
Posts
51 -
Joined
-
Last visited
Content Type
Profiles
Forums
Downloads
Store
Gallery
Bug Reports
Documentation
Landing
Posts posted by lukeoslavia
-
-
So the access log does verify that the traffic is in HTTPS, but it doesnt do this for Roku devices apparently. It only shows HTTP/2 and instead of an address (https://mydomain.com) it shows "-" and then the device name.
any ideas as to why that would happen?
-
I have an emby server running and have setup the letsencrypt docker to get an ssl cert and also to use the reverse proxy. I followed the spaceinvader one video on how to set this up but have not been able to verify that the proxy is actually doing what I want it to do, which is to make sure all incoming requests are over https and all outgoing responses are https.
I have tried checking the nginx access log but that got me more confused than confident as when an initial connection is made, it logs the response as https to my public domain, but after a device begins playback it logs HTTP/2 for the responses and doesn't specify if the traffic is actually being secured(encrypted).
Long question made short, is there a way I can verify that all incoming traffic to my reverse proxy is over https and that all my responses are https as well?
-
4 hours ago, 1812 said:
I have a server with 4 onboard gig ports. I ditched them for a 10gbe card. It's just simpler and faster.
I concur with this, if you have a switch that can do it, you can grab a 10gbe mellanox card on amazon for ~ $35 give or take.
-
You could use the extra nics to create extra networks and separate your traffic i.e. run your VM traffic through one nic, dockers through another, and maybe the third could be for all VPN traffic if you are into that sort of thing? or use them as a fail overs in the event that another one dies.
When I had a Supermicro board, I aggregated all 4 nics to get 4 gigabit throughput which I never really used or needed, but I had it.
I don't think any of these are necessary at all, just things you could do with them / tinker with.
-
On 6/23/2020 at 5:45 PM, keiser said:
Thanks to the brave souls willing to make the jump and test the 6.9 beta. I did end up using my 2 port gigabit card and am running on that, but it'll be nice to get the extra unneeded card out and use the onboard port once 6.9 is released. Hopefully I can get Plex to transcode once the iGPU device shows up as well.
Everything else has been VERY smooth. Frankly I expected more issues and have been pleasantly surprised. Got a week left on my trial and about to buy my license.
I don't know exactly what the difference is, but this also seems to be a gigabyte issue. I loved the i5 enough that I also decided to gift myself a new SFF i5 build for my gaming rig and went with the Gigabyte z490i Aorus Ultra itx motherboard. I had to use wifi to connect and get nic drivers. That's for a windows 10 build too which shocked the heck out of me to see it wasn't already playing nice.
I'm not sure with the ASROCK board in my Unraid machine though as the nic showed up in the current release, but I use a 10g mellanox card instead so it may not have actually been functional. I also saw the same issue discussed in some reviews of their new LGA 1200 boards.
-
On 6/22/2020 at 8:15 AM, nonikka said:
Hello lukeoslavia, could you tell me about the CPU usage when hw transcoding ? for me it takes 40~50% for a 4k h264 96631 kbps video, I thing it is too high and want to know the normal situation, thanks.
TLDR: I have not seen processor utilization that high with transcoding so far.
Unfortunately, I don't have any 4k media in h.264 any more, it's all in HEVC (h.265) now. But I did test a couple of those.
H.265 -> h.264 seems to be the most taxing scenario for the processor. While converting a 1080p 15kbps h.264 down to 8kbps, I also transcoded a 4k HEVC 30kbps video down to 1080p 10kbps and the processor was ~ 17-20% during the entire transcode process for both of them.
So far I haven't seen it break a sweat any more than that unless i'm doing other tasks as well, this thing is kind of amazing for emby transcoding if you turn off the transcode throttle and just let it do the whole video all at once.
The only other thing I can say about 4k transcoding is it's almost always a bear so I limit 4k playback to devices I know can support direct playback.
Also, this is just a wild guess because I know very little about where Jellyfin is these days, but I wonder if their ffmpeg builds are the same as the builds emby uses and if that could cause some extra cpu usage.
-
8 hours ago, FCruz2489 said:
Hi @lukeoslavia, thanks for posting your findings for everyone else to see! I'm getting my 10th gen i7 tomorrow and just wanted to check - did you load the same driver for the 10th gen as previous models? The "modprobe i915" line in the go file or was it something else?
Thank you again
No problem! Yes it is that same driver.
After I set all of that up and then rebooted, I opened a terminal from the webui and typed /dev/dri and it said the directory existed. After that I just passed it through to emby.
p.s. Im loving the i5 10400, its idling at 40 watts and handles transcoding like a dream, let me know how your i7 works out!
-
1
-
-
I've gotten my i5 10400 iGPU passed through using the 6.9.0 beta 1 update. You need the newer linux 5.5.8 kernel to achieve this so if you update and do everything you mentioned previously for passing through your iGPU, it will work (with emby at least).
EDIT: Make sure you have multi-display/iGPU enabled in the bios or this will not work.
-
1
-
-
Answered my own Question! For anyone else wondering how to make this work. Upgrade to the 6.9 beta to get to the linux 5.5.8 kernel (has intel 9th and 10th gen IGPU support) then run through all the other setup needed to make an igpu pass through possible.
Edit go file
Edit syslinux file (in flash section)
Edit docker container to add /dev/dri as a decice.
-
Does anyone know of a way to pass through an Intel 10th gen processors Igpu to the emby docker container? I tried editing my go file and syslinux file and then passing through the device (/dev/dri) to the emby container but this does not seem to work. Im on 6.8.3, is it possible that the newest beta/rc has support that I need?
-
Just now, david279 said:
Yes iCue
Well that worked like a charm, Good looking out! Cheers!
-
The Icue Software ? I will try uninstalling that to see if it helps!
-
I have a really odd issue here. I have noticed while playing apex legends that when I press a key on my keyboard my frame rate drops from the 70s down to ~2-5 fps and the game lags out like crazy, when I let go of the key its back to normal runs like its bare metal right up until i press a key again. I am passing through the usb controller but have always had issues with this keyboard which is why im passing all the usb ports through in the first place. I'd be happy to submit my xml if that would help as well. Im not sure what the issue could be and its about the strangest thing I have encountered thus far.
Setup:
Threadripper 1950x
Corsair Vengance Pro RGB Ram 32gb total / 8gb vm
ASRock x399 Phantom Gaming 6 motherboard
GT 710 (unraid card top slot node 1)
GTX 1080 (Second slot node 1)
Corsair Strafe Keyboard
Ibuypower mouse
Corsair wireless void pro headset
-
Disregard, fixed the problem.
-
I have an 8 TB disk with an ntfs file system that I am mounting outside the array via Unassigned Devices and it houses all of my plex media. Currently I have it mounted in the plex docker with r/w slave as suggested by CA Fix common problems but I cannot write to or delete any files from the disk either in plex or through the smb share. I have a movie that is on the disk twice and want to remove it. I also don't believe that I can record any new OTA shows as plex cant seem to write to the disk.
I've tried so far,
- deleting with mc in the unraid terminal - error this is a read only file system (paraphrasing)
- deleting with filezilla ftp connection - just gives a bad request error
- deleting through smb share - gives no error at all and the file persists
- tried mounting as r/o slave in the docker container - same results
- I also tried running the docker new safe perms in tools but that didn't seem to do anything to the UD disk.
Any help with this would be greatly appreciated, I do not have this data backed up and as it is currently ~7 TB worth of data I do not want to lose it all.
The only solution I can think up is copying it all off to the array which is parity so it will take me a year to do, and then change the file system of the 8 TB disk. I'm not sure if that is even a solution though.
Thanks in advance for any help you can offer!
-
Motherboard: Supermicro X8DTN+-F
CPU: (2) Xeon X5675
Cooler: (2) Noctua NH-U9DX i4
Also comes with 48gb ddr3 ecc ram
Price: $300 USD
-
8 hours ago, bastl said:
The sensors can't be read inside a VM, not that i know of. All the specific monitoring tools like HWinfo or Ryzenmaster don't have access to that. Best solution would be to set up your pump speeds and fan curves directly inside the BIOS.
Just to be clear @bastl, when I check temps I am referring to those utilities in bare metal windows with hyper-v on top. As for the unraid temps I was using the dynamix (I think) temp plugin and reading it from the dash.
@Symon I did pass through mine as well on an unraid Vm, but the temp didn’t read out properly for me, I would check into the dynamix plug in and install Perl with the nerdpack as the temperature readings seem to work well there. Just be aware if you select tdie as the sensor that’s an actual average of the processor sensors not the one that’s +20(ish) degrees.
-
18 minutes ago, jordanmw said:
I never get thermal throttling, so not sure what is going on there, but I did set my pump to 1400rpm- 900 sounds way slow- and unless you have a loud pump, you should get it at least to 1200.
I am curious do know of anyway to control cooler fan and pump speeds in Unraid? I haven't been able to find any info on that.
-
This would be amazing!!! +1
-
Well I can say for sure I am reading the averaged temp from tdie. Nzxt cam, hwmonitor, and the amd ryzen master utility all read from that. I am pretty sure unraid 6.7 rc3 wasnt as it was reading about double the temps im used to seeing. I applied a better coat of thermal paste but even with that, if I have hyper-v running in windows, my temps dont read out correctly and the pump and fan speed wont change at all as load and temp go up. I used prime 95 to peg the processor and the fan speed and pump stayed the same.
With hyper-v not running, the fan speeds go up. I also am idling at 27'C now, but with hyper-v on it idles around 40 best I can tell.
As far as with unraid I never noticed my fan speeds increasing but im not sure if it does or not. I might just have to install it again and test.
-
I was doing some testing with Hyper-v on my 1950x and noticed that when running hyper-v my processor temps could not be read properly and my liquid cooler (kraken x62) was unable to scale fan speeds and pump speeds. Im still unsure what exactly the issue was there but am curious if anyone has noticed this in Unraid.
My processor was reaching temps of 67'C and the fans and pump where still running silent (940 rpm fan speed). This was also causing the processor to lower the clock speed quite a bit to try to keep cool.
So i guess my question is, is this also a kvm issue or just something to do with hyper-v?
-
Thanks for all the tips and sorry for the late reply. I did manage to get my system running much better with your suggestions and the PSA thread was also very helpful!
-
Hi all,
I have a new threadripper 1950x build here and the plan was to run unraid dockers for security cams, plex media, and nextcloud/owncloud with a few vms including one for gaming (windows 10). I ran LSTOPO and made sure I had my cpu's set correctly to be on the same node with my gpu (gtx 1080) but my performance is about 50% and sometimes less than that of bare metal windows. I used 2 ssd cache drives in raid 0 and was gaming off of those and had my os on them.
So my question is, should I expect more than 50% of bare metal performance for gaming and if so how much of a performance hit should I be expecting?
-
Hi all, I have an NZXT Kraken X62 and I am wondering if it is ok or safe to pass-through the USB portion to a windows vm.
I'm not really sure what the USB cable is for and I don't want to ruin my threadripper.
Also, if I do pass that through, will I be able to use the NZXT cam software to control the lighting and monitor the CPU temps?
Thanks in advance for any help you can offer!
[Support] Linuxserver.io - SWAG - Secure Web Application Gateway (Nginx/PHP/Certbot/Fail2ban)
in Docker Containers
Posted
I just set this up a couple of days ago to reverse proxy to my emby server. I am having a strange issue where responses logged for browser traffic show the traffic is using https and where the traffic is going, but for devices (like rokus, xbox, android) it just shows a dash ("-") instead.
For example, this is what is being logged now for devices:
[Devices IP Address] - - [Timestamp] "POST /emby/Sessions/Playing/Progress HTTP/2.0" 204 0 "-" "Roku/DVP-9.30 (deviceID)"
If I connect from a browser the log shows this:
[Devices IP Address] - - [Timestamp] "POST /emby/Sessions/Playing/Progress HTTP/2.0" 204 0 "https://[mydomain.com]/web/index.html" "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/83.0.4103.116 Safari/537.36"
So my concern is, can I customize/fix the log to show the same information for devices as in browsers? I just want to verify that all of the traffic is being sent and received as encrypted https traffic.
Any help/insight would be greatly appreciated here!