bertrandr

Hi All - sorry if this has been asked, I did search and did not see an answer... How can I reduce logging or enforce log rotation? After only a few weeks NginxProxyManager is generating and keeping a LOT of log files. especially in "appdata\NginxProxyManager\log\letsencrypt\" Thanks, BR

Thanks, my little ITX based server board is already maxed out for internal drives, hence the reason I use the great UD plugin. I will soon upgrade my cache drive to 512 or even 1 tb since that hosts most of the busy workloads on my server. Is there anything I can/should be doing to make sure the usb3 connection is using the fastest/ stable config under Unraid? Cheers, BR

Hi All, UnRaid 6.6.6 UD Plugin 2018.12.30a Mounting an external 8TB USB3 HDD - "STEB8000100" Freshly formatted the entire 8TB as XFS I've noticed "lately" (I think since the 6.6.6 update?) that I often loose my UD drive mount. Rebooting the entire server seem to bring it back but it is not an ideal solution. I use the UD mount for 3 main operations. 1.) A target for the latest "CA Backup / Restore Appdata" plugin - backups go into separate folders. 2.) A folder share for my "LibreELEC_Kodi" VM where I store & play very large (40-60GB) UHD/4K movies. 3.) A backup file path / location for my "binhex-rtorrentvpn" docker -mounted with the SLAVE R/W option. I have a single small 250GB cache SSD where most of my active torrents seed from. However when seeding several large torrents I sometimes move them to the UD folder to free up space on my SSD. My UD mount can have anywhere from 5 to several hundred open files - depending on the Torrents I might be seeding. In general performance from the UD mount is awesome. No problem streaming 50GB 4K MKV's to the Kodi VM. While simultaneously seeding a few torrents. My question is: Is there a practical limit to what sort of load I should expect from a decent 7K SATA USB3 drive mounted via Unassigned Devices? Is having a few hundred open files across the USB3 link too many? -And maybe causing the plugin to crash? In general if I leave it alone, UD seems fine. But If I do a large file copy to/from from my W10 PC it seems to suddenly dismount and things go sideways... Thanks for all the great work. Bertrand.

Fantastic work! Followed instructions to a "T" and now my moderate home network is add free!!! Thank you, BR

Great docker! Thanks for the hard work -I much prefer over CouchP... Questions: 1) Can an Indexer be weighted or preferred. When two very similar results are found from I would prefer to download from my primary private tracker (IPTorrents) 2) If I have my DL client (ruTorrentVPN docker by Binhex) configured to "move completed DL's via Hardlink" -which copy does Radarr import from? The original (still seeding) location in my /incomplete folder, or the moved & completed copy in my /completed folder? Ideally after Radarr grabs a movie I want to move (hard-link) it off my smallish SSD cache drive to a "completed" share on UnRaid. From there import the movie into Radarr & my library... The goal is to download active torrents on the SSD cache drive them move & continue seeding the completed torrent from a regular UnRaid share. (Beer fund donation sent!) BR

I noticed there is a new version - any specific updates? FYI - I have started to use the Web GUI -much easier than the VNC session. BR

SMTP settings? Maybe I'm just blind, but where do we put the SMTP server for email notifications?.... BR

Bad news - changing the advanced settings makes little / no difference. Backups still "stop" randomly. I've opened another ticket with CloudBerry and they quickly resonded: "Hello Bertrandr,Thank you for reporting. That is a known issue. We are working on the fix and will let you know once it's ready.We apologize for inconvenience caused." I'm still on the trial version (which I like), but until this issue is fixed I'm not buying... BR

Yea my backups to B2 seem to randomly stop as well. I was hoping this recent update would fix it but my backup stalled again this morning. It seems to happen with very large files over 1GB (home video archives). I have since been tweaking the options under the advanced settings. I read somewhere that Backblaze B2 stores in 100mb chunks or blocks and the recommendation was to try and match the incoming "chunk size" from Coudberry to that of Backblaze... But you also need to make sure that CloudBerry has enough RAM assigned to support the number of worker threads and chunk size... The formula to calculate the minimum RAM allocation is <threadcount*2*chunksize> Personally I am using 3 threads, 100mb chunk and 700mb RAM allocated. Read this: https://www.cloudberrylab.com/blog/slow-backup-how-to-speed-up/ So far so good, I have been running a very large backup (87 GB) with many large files for over 3 hours without stalling. I'll report back tomorrow... Cheers, BR

Forced an update to pull down the latest image without any issues. Still using the CloudBerry trial key and Backblaze B2 storage over my 15 mbps (upload) internet connection (Shaw 150 unlimited) Test backup is a 33GB folder on my Unraid with a couple thousand JPEGs Update looks VERY promising!!!! CPU utilization is negligible (5-6% overall) even with encryption enabled -as it should be. Upload speed is maxing out my internet link at a steady 1950+ KB/s (1.95 MB/s) Running for 20 min with little fluctuation in CPU or upload speed Next test will be to see if it stalls after an hour like the previous version. I'll report back when the test finishes... Thanks for the quick turnaround Djoss! BR

Not sure if it helps you, but I just heard back from CloudBerry support - apparently the <just posted today> Linux version 2.1.0.81 has a fix?

Yes I do, turning encryption on/off made no difference. I also experimented with enabling / disabling the bandwidth throttling feature but it made no difference either. -Restarted the docker and backup job after each change. Without throttling, encryption or compression -I'm not sure why the app is so CPU hungry if all it is doing is copying jpeg files into Backblaze. I have opened a ticket directly with Cloudberry and apparently it is a "know issue"? and a "fix is on the way" ? I'm open to suggestions and willing to try other ideas! I know I could pin the Docker to a single CPU thread, but that seems more like a band-aid, I would rather help find a fix. Cheers, BR

Anyone else seeing very high CPU utilization during backup? I have a decent quad core E3-1220 v3 @ 3.10GHz in my server and this CB docker is pushing all 4 cores to 75-100% while running backups. I have tried adjusting the Threads, Chunk size and Memory settings but they don't seem to make much of a difference. My current "test" backup is mostly jpg files, encryption is on, compression off -backing up to Backblaze B2 buckets. Bandwidth throttle set to 1024 KByte/s on a 15 mbps upload intent connection. Cheers, BR

My 2c, Sometime in the past 24hrs Deluge VPN stopped being able to resolve "nl.privateinternetaccess.com" (discovered with DEBUG logging enabled) -Without it, Deluge VPN won't properly start and allow access to the Web console unless you turn the VPN function off. I ended up hard coding the current IP for nl.privateinternetaccess.com as 109.201.135.220 and everything started working. I consider this a temporary fix, as the name resolution is critical when the NL PIA gateway changes IPs.... I wonder if this had something to do with the DDoS attack today on some of the Internet's DNS servers?... I actually like the fact that Deluge VPN shuts down when it can't maintain a secure VPN - better safe than sorry! Cheers, BR

I have the same problem running Chrome on Win7 - cannot control the cursor <with my mouse> over the CrashPlan app in the VNC window... ("Local Cursor" is grayed out within the VNC settings) But using IE on the same PC it works fine... (Local cursor is enabled within the VNC settings) I've isolated the issue, it seems to be related to my laptop having a touchscreen. I can "touch" the CrashPlan app via noVNC just fine using my laptops touch screen using Chrome browser, but soon as I use the mouse it stops! Something about using a mouse on a touchscreen enabled device and noVNC over the Chrome browser... breaks. My 2c.

+1 This type of solution is very much needed. Having the ability to choose which application uses the VPN provides a great amount of control and flexibility. As an example VPNing docker applications like Sickbeard, Sickrage, Sonarr, Couchpotato, Sabnzbd, to name a few, but not VPNing Plex Server since it has an issue with losing it's connection once a VPN connection is established. Hope to see this type of solution come to fruition in the future. OR... You can use your external firewall / router. Load a DD-WRT or Tomato firmware on your router and using the OpenVPN client & the active routing policies - you can "route" any number of specific internal source IP's out through the VPN. Keeps the networking on the edge and it is actully quite easy to set up (If I can do it! ;-)) My 2c,

im not sure which is cheaper but i can tell you PIA are pretty darn cheap if you go for a year, i think im paying around £3 per month for unlimited usage, speeds are ok, not lightning fast but adequate for me (approx 800 KB/s to 1.3MB/s DL on a 20Mb/s line) +1 for PIA, I am in Canada and pay $6.96 USD / month for PIA. Speeds are great, my current PIA endpoint for Deluge is in the Netherlands and I still reach 2-4 MB/s (Bytes NOT bits!!) on well seeded torrents. I also create a separate tunnel to the US on my Tomato Router for Netfilx using sourced based routing and openVPN -also fast and reliable! Downloads caps (800 GB/mo) are set by my ISP - PIA doesn't care... My 2c BR

May I ask which specific brand GT730 (passive/active) you are using?

I have very similar issues -tried using both ASUS GT720 and a ASUS GT610. Everything would be fine until the first reboot -then it would lock up my entire UnRAID server (running UR 6.1.3). I am waiting for the next release / update to OE before I try again -I don't have much time right now to tinker with different configs. I only want to use small & passive GPU's due to the size of my case & where it lives in my house. I'll spend the money and test another card if there was good experience with it, but it needs to be a single wide & passive GPU. Oh and it needs to support bit-streaming HD audio through OE... I have not tried the "MSI interrupts enabled" fix yet, but i have confirmed that my GPU does not share anything unusual within the IOMMU Groups. My 2c BR

Yep! -You need to enable the passthrough mode and select the correct audio device. In my case it is the second selection in the drop-down audio option -the name of my pass-through card + "Samsung" (my TV) -this is what my receiver is reporting back through the HDMI EDID handshake to KODI. KODI via the VM OpenELEC can "see" the same downstream components & GPU as my standalone OE system, but for some reason not send the "pass through" bit-streamed HD audio through the GPU card in my UR server... What MoBo are you using? Have you tried playback of DTS-HD or Dolby TrueHD on your virtual OE -assuming your receiver supports it... BR

-I agree & totally understand. A verified HCL would be great, one of the reasons for my long post ;-) Yes -passthrough audio on my standalone (OE Stable 5.0.7) system works perfect with either the GT610 or GT720 -audio is rock solid! I have not tried the OE beta's yet since I was happy with the current V5. The Latest, Beta 3 OE Stable 5.0.7 I would agree! & no I left everything for machine type at the template defaults = Q35 2.3 You want me to try a different version of Q35 ? Sounds reasonable, what is the best way to override the VM resetting behaviour? -if any... If by "issue" you mean compatibility I would agree. I am very confident that there are no HW errors/issues with my server. Every component was thoroughly tested before I installed UR. MoBo: AsRock E3C226D2I, CPU: Xeon 1220v3 (4 Core) Cheers, BR

First off -thanks for all the awesome work that has gone into UnRaid, especially the VM capabilities! I've been waiting to try out the OpenELEC template/image for quite a while. Hoping one day it could replace my standalone OpenELEC HTPC. I am running a relatively new HW build (see sig) and UnRaid 6.1.2 I have never had a single problem upgrading or with overall system stability & functionality. This weekend I finally did some testing with the current OpenELEC template -my results below. Creating the VM with passthroughout GPU and HDMI sound was easy. I tried two different passive GPU cards in my UNRaid server, an ASUS GT720 and a ASUS GT610. The GPU's were connected to my Yamaha receiver over HDMI which is connected to my TV over HDMI. Note -Both GPU's work PERFECTLY in my dedicated OpenELEC HTPC (version 5.0.7) -also connected to the receiver... Both seemed to "pass-though" OK and worked for the initial "create & start-up" using all the default settings in the OpenELEC template. I was also able to connect / add my IR receiver and wireless keyboard! Initially everything seem great! The GT720 picture quality and repressiveness was good, until I tried to enable pass-though audio mode in KODI. There was no sound at all... Turned off -passthrough and audio returned. Twice while testing KODI menu sounds did come through in audio pass-though mode, but it was "stuck". Like a button press being held down and echoing over and over until I turned pass-through off. I have allot of movies & BlueRay rips with full DTS-HD and Dolby TRU-HD formats that require HD audio pass-through to my receiver. -Again my baseline is the standalone OpenELEC system that works & sounds perfect with the same GPU. I tried everything again with the GT610, picture quality and repressiveness was also good. This time pass-though audio (menu sounds) worked right away! -Until I tried a movie... Mad Max Fury Road has a fantastic native DTS-HD sound track! but playing through the OpenELEC VM with pass-through audio enabled was full of audio drop-outs. Randomly every few seconds or minutes -the sound was dropping out :-( my baseline standalone OpenELEC system plays this perfectly. I tried using SEABIOS and OVMF but the audio problems were the same. I tried using a dedicated mapped video.rom via the XML template -same issue. Standard audio works great, but trying to pass through HD audio failed. BUT, this is not the worst of it. After rebooting either of the VM's (from within KODI) my entire unRAID system would appear to lock up! and the OE VM would never fully boot. I think I traced the problem to the PCIe / Network stack. If I ran a continuous ping to my UR server while rebooting the OE VM, pings would fail just after the OE splash screen appeared. My UR server would be totally fubar'd and the OE VM never managed to reboot fully. My direct USB keyboard connected to the UR console would also freeze! The first time I left it for an hour, but was forced to do a hard reset on my UR box to recover! I started to tinker with the OEvm network settings (br0 or vibr0) but the same result each time -my entire UR server would lockup after the OEvm rebooted for the first time -after the initial build & boot. I tried changing the "PCIe ACS Override" settings, setting static IP's in the OEvm -nothing made any difference. I think something about passing through the GPU or USB to the OEvm in my specific HW mix, causes my host hardware to loose control over it's entire PCIe / USB / on-board network stack. Only a hard reboot can recover. Testing all this took a LONG time and forced me to hard reset my UR server more than 10 times! I will gladly send anyone my config details, but a full retest will have to wait a few days until I get some free time... Cheers, Bertrand.