Ezro

Hi all, It takes 15+ seconds to load Movies in Jellyfin and 3h to 'Refresh Movies' in Radarr. I've been trying to identify the bottleneck for weeks now and still can't pinpoint why I'm experiencing this slowness. Can someone please help vet my unraid configuration?

I meant to disable the FileBrowser login screen

@alturismo Do you know if it's possible to allow public access to the FileBrowser? Right now I have most of my app access managed via nginx and it'd be nice to have this do the same

Ty for the info! I completely misunderstood this. I'll look into webdav for accessing / syncing my unRAID share via a public (reverse proxied) URL

I was able to set this up and have it working behind my reverse proxy (e.g., https://filebrowser.mydomain.com). If I use the browser and go to the URL, I'm able to log in and upload / download files. If I wanted to use the FileBrowserGO iOS app though is there a way to add it as a remote location that way Photos can be uploaded / synced from my phone without having to use Safari (or a browser)? Or similarly, if I wanted to directly map this on my Macbook, is there a way to do that without having to use a browser?

Hey everyone, I'm running into a bit of a pickle and I'm hoping someone can help with my configuration. My current setup consists of: Gandi domain + Cloudflare proxy providing SSL cert => the old version of this docker container to provide the reverse proxy Duckdns subdomain => the old version of this docker container (letsencrypt) to provide the reverse proxy and SSL cert I've recently removed Cloudflare and am now running into an issue where my domain is using the cert of my duckdns subdomain. I don't mind upgrading the container but would anyone be able to provide insight into how I would need to configure the parameters to accomplish this? (Or am I better off leaving the existing deprecated container and trying to somehow generate a new cert for my Gandi domain?)

@JorgeBI ordered a replacement extender backplane (https://www.ebay.com/itm/274829421125) which is the same model as the one I replaced. After swapping it out, I've been testing it and seeing the same issue(s). Sometimes I'll restart it and no SCSI devices appear in the Tools->System Devices->SCSI Devices. Sometimes a few drives appear. Sometimes almost all (but one) drive appears. I haven't yet had the new backplane show all drives in the SCSI Devices pane. Do you know how I can determine if maybe I'm not powering all of the drives well? Or maybe it's the cables I'm using? I'm not sure how to diagnose what's going wrong but I'm going to switch back to the original backplane to try to get all SCSI devices recognized.

A bit of a newbie question, but is there an easy way to detect how many / which drives are being detected? (Preferably by serial number)

@JorgeB I still have this issue pretty much every time I restart the server and it usually takes a bunch of reboots in order for everything to start working again. Last year I ordered an LSI SAS 9210-8i and swapped it out, but that didn't resolve the issue. I currently have a Supermicro 846E1-R900B server and am not sure how to diagnose if it's something with the SAS backplane or if I happened to receive another dud HBA card. Do you have any advice / steps on how I can diagnose this and hopefully fix it?

For anyone interested: I changed the Radarr config path to /mnt/cache/appdata/radarr and that sped things up dramatically (down to under 20s). I was advised to also move my Docker to the cache (Settings -> Docker -> Docker vDisk location and Default appdata storage location), but did not notice any additional improvements from that; it could be that the unRAID container updates and restart are faster now, but I don't really notice.

Hey everyone. I've been noticing performance issues for a while now after updating to the new UI (v3?). It currently takes over 3 minutes to load the Movies section each time I refresh the UI. I was thinking it may be because the container wasn't using my SSD cache for the config; initially my appdata User Share was 'Perfer' for Cache, and now I've made it Only. I've ran the data mover and have restarted the container but am still seeing the same slowness. Currently my container is set to use: /config <-> /mnt/user/appdata/radarr Could it be that that is still somehow using the spin disk instead of the cache? (Should I update it to /mnt/cache/appdata/radarr? I've seen threads mention that this is a bad practice and /mnt/user should be used)

Thanks for the reply Squid! With this approach, is it possible to expose both of the storage through the main unraid server as a single network share? I don't mind adding additional path mappings for the docker containers, but for PCs on the network I'd really like to expose it as a single storage that way people don't have to worry about which network share has space / where their content is, etc.

Hi everyone, My setup currently is a single server with the following: Docker Containers * Radarr * Sonarr * Transmission * Nextcloud Shares * media - used by Jellyfin server as a mapped network drive * share - used by PCs on the network for misc. storage, usually mapped as a network drive * recordings - used by an OBS machine and editing machine as a mapped network drives I'm expecting that we may run out of space on the server by the end of the year so I want to try to plan out how to expand the storage properly. One option would be to set up another Unraid server and basically duplicate the above configuration(s), but I don't like that solution since I'm not sure how radarr / sonarr would know about what's on the other server, same with user experience for exposed shares (i.e., "where was I editing my clip again?"). Does anyone have advice on how to expand storage that way all of the storage is seen (by the docker containers and machines on the network) as a single pool of storage?

Hi everyone, I currently have a working LetsEncrypt (port forwarded 443 -> 444) + DuckDNS for all of my docker containers hosted on my unRAID. I added a new machine to my network that is running Windows 10 and hosting a Jellyfin instance. If I want to use a different domain (not DuckDNS) for Jellyfin, how should I set things up that way I can keep my existing DuckDNS + my new domain all as HTTPS? Edit: I was able to get it working by pointing my CloudFlare DNS to my public ip, and creating the appropriate proxy-conf with the server_name set to my non-DuckDNS domain. If I understand correctly, this means that my SSL isn't actually being created / handled by the LetsEncrypt docker and is being managed at CloudFlare. Does anyone see any issues with taking this approach?

I'm able to connect to GitLab and it appears to be running correctly, but an error keeps spamming my logs: ==> /var/log/gitlab/prometheus/current <== 2019-12-04_06:09:35.37530 level=info ts=2019-12-04T06:09:35.375Z caller=main.go:293 msg="no time or size retention was set so using the default time retention" duration=15d 2019-12-04_06:09:35.37547 level=info ts=2019-12-04T06:09:35.375Z caller=main.go:329 msg="Starting Prometheus" version="(version=2.12.0, branch=master, revision=)" 2019-12-04_06:09:35.37555 level=info ts=2019-12-04T06:09:35.375Z caller=main.go:330 build_context="(go=go1.12.13, user=GitLab-Omnibus, date=)" 2019-12-04_06:09:35.37563 level=info ts=2019-12-04T06:09:35.375Z caller=main.go:331 host_details="(Linux 4.18.20-unRAID #1 SMP Fri Nov 23 11:38:16 PST 2018 x86_64 2cc6f067be8c (none))" 2019-12-04_06:09:35.37570 level=info ts=2019-12-04T06:09:35.375Z caller=main.go:332 fd_limits="(soft=40960, hard=40960)" 2019-12-04_06:09:35.37575 level=info ts=2019-12-04T06:09:35.375Z caller=main.go:333 vm_limits="(soft=unlimited, hard=unlimited)" 2019-12-04_06:09:35.37916 panic: runtime error: slice bounds out of range Has anyone come across this before?

I decided to give this a go and was able to successfully mount to my NFS share for Plex. In doing some testing, though, I'm noticing that both SMB and NFS have pretty bad (relative) scan times when it comes to my Plex libraries (around 2-10 Mbps in my Windows network throughput graph). On my Windows machine (PMS), I've tried changing the mount to use nolock and have also tried TCP / UDP / TCP + UDP for the NFS client. On my unRAID server, I've tried Auto / Yes for the the Tunable (enable Direct IO) setting. The machines are connected via ethernet on the same network. My shares are running using 5400 RPM 6 GB/s spin disks, and my Plex library is structured like this: /mnt/user/plex/movies/{movie_name} (movie_year}/{movie_name}.{ext} Using Crystal Disk Mark, I'm able to get decent synthetic scores (30MB/s NFS vs 63MB/s SMB sequential read), so I'm not entirely sure what's causing the slow scanning. (I also was expecting NFS to generally outperform SMB, but it was only better in random read / write, while SMB was better in sequential) Does anyone have any advice on how I can beef up my connection so that I'm able to push more than a few Mbps?

It doesn't look like it. I tried switching the port that the HBA card is plugged into and that also didn't work. I'm not sure if the power outage possibly fried the card itself or if it's my motherboard's ports. It'd be weird if an unused port was fried out, though, so I'm more inclined to believe that it's the HBA card. Do you know if there's possibly a bios property that could have been switched if the bios reset?

Hi everyone, Last week my server shut down unexpectedly and I've been working on fixing it; after bringing it back up I'm seeing all of the disks as missing except for the boot device. I have a feeling it may be something with the SAS HBA, but it's been a while since I've done any work on this server. If anyone has any ideas / tips to try to get the OS to recognize the drives again it'd be greatly appreciated. Attached is the diagnostics. Regards, Ezro tower-diagnostics-20190806-1738.zip

I disabled all of the other devices in the bios (not just the boot order) and that did the trick! Now it's time to rebuild all of the settings and containers. Sorry if I came across brash, this has been causing me a lot of stress. Thank you everyone for your assistance.

That's fair. I'm able to boot into FreeDOS if I unplug my LSI card, which leads me to believe that the LSI card is the problem. Because of that I'd like to flash my LSI card and remove the bios, but I don't know how to boot into FreeDOS with the LSI card installed. I tried taping the pins, but that didn't work.

I feel like I'm missing something for my understanding. I'm still trying to boot into FreeDOS with my USB. In my bios, I disabled every boot *except* for my FreeDOS USB, but it's still booting into MPT2BIOS-7. I understand that I want to remove the bios from the LSI card completely, but I'm not quite sure how to do that when it's not respecting the BIOS boot order to begin with. I've also tried electrical taping pins 5 and 6 (as per the SAS HBA crossflashing article), but I'm seeing the same results. Also, from the other previous link, the outcome I'm expecting is to flash the firmware to IT mode, but is that the same as removing the bios? If so, my LSI card is already flashed to IT mode...

@jonathanm @johnnie.black I'm starting to look back into this and am noticing that I'm still not able to boot into my FreeDOS USB in order to attempt to flash the LSI card. In running the configuration utility, it does look like my card may have been pre-flashed when I bought the server. I've attached a picture showing the card's firmware which shows it running 20.00.07.00-IT. Do you guys have any other advice on what could be the issue?

Thankfully I labeled my drives on the physical chassis, so the assignment shouldn't be too bad. Also, I don't have a backup of the diagnostics zip or anything, but I still have the email for my Pro Key; hopefully all I have to do is restart using the new SanDisk USB and enter that into the web UI.

Ah, I see. Thanks.

Thanks for confirming. After I flash the LSI card, I'm afraid that my OS stick is already borked from Windows repair; if I have a new install on another USB, will it recognize the drives and the data? (Does it matter what the sequence is? Boot into OS, restart with LSI vs Boot into OS w/ LSI on first start)