Chamzamzoo

I see, that certainly is clever Opening the ovpn files for the app and the one I uploaded to app-data, they are both using aes-128-cbc and other the settings are the same. What do you do with the incoming and outgoing port settings in Deluge, they need a value? Set it to random? or is it irrelevant whatever you put in it?

Could it be the type of VPN that VM is slow with? Using the OS X PIA app with killswitch on and IP checked to make sure, I DL'ed that file above at 5MB/s using the PIA client. Changed to Czech Republic and it was MUCH faster than France. Not sure if the app would be using OpenVPN or not.. Changed the Deluge VPN to Czech republic and restarted Deluge, just as slow. Or is it a port issue? When you change locations with PIA you get a new forwarded port. Do you put this in the WebGUI Deluge preferences incoming and outgoing ports or in the Docker settings? Also couldn't get wget to work: Total Download Size: 0.61 MiB Total Installed Size: 2.81 MiB :: Proceed with installation? [Y/n] y :: Retrieving packages... error: failed retrieving file 'wget-1.20.1-3-x86_64.pkg.tar.xz' from mirror.pseudoform.org : The requested URL returned error: 404 error: failed retrieving file 'wget-1.20.1-3-x86_64.pkg.tar.xz' from mirror.pseudoform.org : The requested URL returned error: 404 error: failed retrieving file 'wget-1.20.1-3-x86_64.pkg.tar.xz' from archlinux.dynamict.se : The requested URL returned error: 404 error: failed retrieving file 'wget-1.20.1-3-x86_64.pkg.tar.xz' from arch.mirror.far.fi : Protocol "rsync" not supported or disabled in libcurl error: failed retrieving file 'wget-1.20.1-3-x86_64.pkg.tar.xz' from mirror.pseudoform.org : Protocol "rsync" not supported or disabled in libcurl warning: failed to retrieve some files error: failed to commit transaction (download library error) Errors occurred, no packages were upgraded.

Linksys Velop mesh network. Going to see if I have any success in modem & router mode.

See I was fine the PIA on BT infinity, practically full linespeed for torrents. Only thing I've changed is moved to virgin media and now it sucks. Limited to ~900KB tops. Tried all the above, router is in modem mode.. don't really know if there's anything else to be done other than ditch VM.

did you have more success with a different country, or just move to Nord? Have you managed to configure Binhex Deluge with Nord? that's my next thing to try.

It was completely crippling the rest of my conneciton when DLing a torrent, speedtest.net wouldn't get a ping to begin the test, testmy.net was reporting 2Mbps.. I did find this: https://www.privateinternetaccess.com/archive/forum/discussion/23112/anyone-having-torrent-slowdowns-please-read-this Mushroom's post about Deluge and the ITConfig plugin to unset incoming and outgoing utp has helped the rest of the connection become uncrippled, but the torrents are still slow. I am using the hub in Modem mode... my Velop router is taking care of the rest.. just outside my cooling off period as well.. Did you get a new Modem or still using the superhub? I am in Modem mode at the minute and it's not really helping.

Just changed ISP to virgin Media from BT infinity... Speeds seem to be capped at about 7-800KB now, anyone else epxeicned this? Is this a limitation on Virgins part? Using Switzerland as my PIA destination with port forwarding set up, although I can't be sure this is working. Is there a way to confirm the port is indeed open?

anyone have issues with out of memory errors and the Nvidia control panel not being availabe until reinstalling drivers after booting between the VM and SSD? In device manager it looks like the driver has also rolled back to 388.59 from 2017 when I just (re)installed 417.35 from this december afterbooting between the SSD directly and the VM version. Very odd.

In the guide SI1 said that he shrank the volume to make use of the unused space for Vdisks.. I don't want to use vdisks, but just wondering how he was intending to have the NVMe drive accessible to unraid while it was passed through. On my system it vanishes, it's not available in unassigned drives or anywhere until you remove it from the vifo driver on the flash USB disk. Not a big issue, everything works as expected. just you lose access to any of the NVMe from unraid unless you're booted into it.

Brill guide, again just what I was looking for. I have a question, in your video you mention that you're keeping half the NVMe for vdisks, which I assume must be accessible by unraid? When I pass through the NVMe via the vifo driver stub, I lose access to it in unraid. Am I missing something?

+1 yes please

Thanks for all the explanations, it's coming together slowly in my head. I'm not sure how I would know which core has access to the memory though, any chance would you be able to explain that bit for me?

Thanks @bastl, that's very helpful. I'm mostly interested in performance for the gaming side of things, my plex and other dockers don't see a huge amount of usage. Would gaming benefit from either increased bandwidth or latency, or is it even individual game dependant? I will try the Epyc code and see how it runs in some benchmarks, looks interesting. I tried this as a more traditional main core + HT core assignment but it was back to dual channel mode I think, scores were down 50% for L1-3 cache. Sorry for the large images, they come out like this on Mac.. must be the retina screen.

I've been following this topic with interest but can't say I understand more than half of it yet. Things are a little different for me with threadripper 1900x, I've set my CPU's as follows: This did deliver a marked improvement in AIDA64 over my previous setup (which had a 6 pairs for 12 cores): New setup: Not entirely sure how the NUMA settings would affect me on this CPU as it has less cores (and perhaps less dies. Would it be worth changing the BIOS setting to Channel as well, considering that AMD "confines the 1900X's active cores to a single CCX inside each die"? I see some comments above suggest that any of this may not actually make any real world difference, or am I misreading? I did get a bit lost after page 1. The AIDA64 numbers do look markedly improved though!

Well I fixed my issue but unfortunately it was a threadripper specific issue. There was a “dirty patch” to make threadripper work with vm gpu passthrough which became irellevant after a uefi motherboard update, so it was removed. I hadn’t done this update, and had no idea it was even there (could’ve had a warning pop up at least..) perhaps it’s something similar for you?sorry can’t be more help.

No reply? thats concerning. I have the same issue (although on threadripper) and keep rolling back to 6.5 which does work again for me.

Hello, I'm wondering if UnRaid can solve a challenege in my work life.. We have a number of students who need to be able to host HTML resources they produce, on public facing webspace. We use an active directory controller to handle their logins. Do you think (with a combination of AD & Web server dockers / plugins) I would be able to achieve a fairly simple system where students (using AD credentials) log in to their own area (via browser or mapped drive), upload some HTML files and retrieve the URLs to deploy their resources to the wider world? We currently use a Synology box for this (long story), and while it works, it does create some awkward URLs that have ~DS%5c in them. This is butchered by almost any email client or HTML editing package into an incorrect URL (%5c is code for "\") and it is causing us all sorts of headaches.

Edit *** it's just started working again *** After your reply I can see that your Deluge docker is working properly, but my system settings are not being respected by the browsers, despite being told to use system proxies in settings (chrome and FF). When I put the proxy into Firefox directly, it works, but when set to system settings (and proxy applied to the wifi device) it doesn't work. It has worked like this before. Weird! My dockers use it also (sonar etc), and for a time they were not getting any data through for searches (how I noticed the problem). They are now (not sure if I can check over what IP though).

Proxy doesn't appear to be working again, any ideas? Downloads have the VPN IP but proxy just goes straight through to my ISPs IP. supervisord.log

Would that have meant that if a disk had failed while those errors were present there would have been problems recovering from parity?

Ok I’ll kick another one off and see what it says. Thanks. edit: Zero errors now

Hello, Scanned a few threads I could find on this and couldn't find much, can anyone advise? I've ran the parity twice now and got the same amount of errors (1687), there have been no unclean shutdowns to my knowledge and the SMART status' appear to all be fine. Attached are two diagnostics, one from after each of my recent parity checks. Thanks for any advice. repeated check still 1687 errors - tower-diagnostics-20180327-1641.zip after 1st 1687 errors - tower-diagnostics-20180326-2040.zip

After the update last night I no longer have the proxy issue Thanks Binhex and Strike for helping.

You're right. I had assumed it was in an error state as it paused my downloads and repeats this message every 30s, it looked like it was retrying for a port. Downloads are running, but I don't get a PIA IP via the proxy anymore (another thing that led me to believe it was down). Any ideas? supervisord.log

I've been on NL from the get go and can't get a port either. Happened in the last day or so.