ConnerVT

Reading this thread, I think it's time for a quick Docker tutorial: -- the docker.img file (which typically resides in the Unraid system share on the cache pool drive, by default) is where the docker "programs" reside. -- each docker program's configuration files, logs, databases, and other support files are typically stored in the docker's appdata, which usually is in a folder in the appdata share (usually on the cache pool drive, for fast accesses). This is typically mapped from inside the docker container to the appdata share folder by a Container Path entry in the docker's template. -- For dockers which manage a large amount of data (media servers, photo programs, torrent/NZB downloaders, etc.) data is typically stored on the Unraid array, and the docker has a path mapping from inside the docker container to where the data is in the array. In Unraid's Dashboard System tile, the Docker line displays how much space the docker programs are using of the allocated docker.img file (Unraid's default size for this file is 20GB). Ideally, the only time this size should change is when you install an additional docker program. However, it is good to monitor this size, as sometimes things mess up, and require you take some action. Some possible things that happen are: -- Improper Container Path mapping - Say a download program does not have a Path mapped outside the docker container. Downloads will be written to the docker.img file instead, quickly filling it and causing all of your dockers to crash. -- Some containers keep their log files with their program in the docker.img. Check if there is configuration setting in the program to rotate out old logs, and avoid using debug logging if unneeded. -- Almost always avoid updating a docker from within the docker, better to update the entire container. Some programs store backups when updated, which will likely end up in docker.img. I have 37 Docker containers installed in my server. 25 of them running all of the time. My docker.img file is at 22.9GB used.

I believe it to be an Ultrastar He8 - One of HGST (Hitachi) first attempts of a helium filled drive (He6/He8).

First thought is that you have some Plex appdata on your array. Quick way to check is click "Compute" for your appdata share.

@xokia I haven't used NUT. I believe the main benefit is that it allows multiple devices to run NUT (one as server, others as clients) so you can power down these devices as well if running on UPS power during an outage. apcupsd may have this functionality as well, but I haven't investigated this as I currently don't have a use case for it. You may wish to reduce your run time before initiating shutdown. A UPS should not be run down to less than 50% battery capacity. It is very hard on the battery, reducing battery life. It also takes significantly longer to charge a battery than discharge - 10 minutes of run time could take hours to return to 100%. Power outages also aren't consistant. Power could go out, come back, then go out again. The "clock" restarts when power comes back on, but now you may be starting at 90% instead of 100% (even after just a minute). You also need to take in consideration that it takes Unraid a bit of time to do a clean shut down (the goal for a UPS protected system). Unraid spins up all drives, stops all running Dockers and VM, completes any cached writes in RAM, spins down the array, shuts down any remaining processes, then powers down. This may take several minutes, depending on what you have running on your server. It is wise to run a shut down (maybe a few times) and time how long it takes. Use this information to decide how long the server should run before initiating a shut down. Here are my settings. I let the server run on UPS for 6 minutes (Time on Battery). If the battery level falls below 50% or the UPS says it has less than 12 minutes of runtime left I shut it down (this handles power on/off/on/off situations):

@kiwijunglist You may need a new battery. 5 minutes of run time reported is awful low for 100% battery on a 900W UPS:

Unraid supports apcupsd (look in Settings) natively as well as a NUT UPS plugin.

I'm not having any issue, but I also see this GPU not available bound to VFIO or inuse in a VM message when I expand the GPU pane in my dashboard. It only pops up for about 2 seconds, then disappears. I have both a Nvidia Quadro P400 and a AMD 3700G iGPU active (both configured with appropriate driver/TOP). I only configure to display the Nvidia stats with this plugin, which works just fine. Neither is bound, nor passed to any VM. Running Unraid 6.12.8 and the latest GPU Statistics 2024-03-16. Just thought I would pass along this data point.

Run less Docker containers? All this should fit easily in the default 20GB dockerimg file. If it isn't, you likely have something mapped incorrectly, where it is writing in the dockerimg vs to an outside drive/pool/array. This fits well in a 30GB dockerimg: Name Container Writable Log immich 4.16 GB 2.04 GB 1.93 MB binhex-krusader 2.58 GB 0 B 52.4 MB PhotoPrism 1.82 GB 0 B 52.4 MB binhex-delugevpn 1.62 GB 949 kB 942 kB binhex-sabnzbdvpn 1.25 GB 100 MB 13.6 kB homebridge 998 MB 85.6 MB 11.4 MB DiskSpeed 869 MB 1.74 MB 21.8 kB Thunderbird 807 MB 7.35 kB 7.89 kB ApacheGuacamole 795 MB 57.8 MB 89.6 kB overseerr 702 MB 2.22 MB 16.0 MB PostgreSQL_Immich 687 MB 63 B 5.41 MB homarr 663 MB 0 B 52.4 MB plex 615 MB 275 MB 213 kB dupeGuru 569 MB 0 B 52.4 MB nzbhydra2 530 MB 0 B 52.4 MB nextcloud 448 MB 597 kB 1.36 MB bazarr 428 MB 23.2 kB 3.27 MB bazarr4K 391 MB 73.3 kB 41.9 kB speedtest-tracker-1 343 MB 27.1 MB 679 kB sonarr 335 MB 138 MB 18.8 MB mariadb-next 333 MB 25.2 kB 8.33 kB QDirStat 314 MB 19.7 kB 33.7 kB gaps 300 MB 32.8 kB 88.9 kB lidarr 270 MB 23.6 kB 1.10 MB nzbget 208 MB 27.9 MB 7.31 kB radarr 198 MB 23.6 kB 3.41 MB radarr4K 198 MB 23.6 kB 1.19 MB tautulli 197 MB 23.2 kB 483 kB SpeedTest-By-OpenSpeedTest 181 MB 4.73 kB 40.7 kB NginxProxyManager 178 MB 48.4 kB 6.72 MB jackett 171 MB 0 B 52.4 MB Redis 138 MB 0 B 997 kB glances 73.8 MB 0 B 52.4 MB Rickroll 72.3 MB 5.06 kB 1.17 MB cloudflared 51.3 MB 0 B 2.70 MB Docker-WebUI 17.5 MB 0 B 44.0 MB Dozzle 15.5 MB 0 B 776 B Total size 23.5 GB 2.76 GB 489 MB

You posted a link to a post I'm confident the OP knows about (who doesnt at this point), which points out the very recommendations they're saying are no longer available, from a test two years ago. How helpful 👏 If you read through the thread, you will see it is more about the current state of the USB Flash drive market. There has been a large number of counterfiet flash drives flooding a number of the major Internet marketplaces, making it impossible to just say "Buy this brand and model...". As I've recommended several times in that thread (and in countless others, all asking the same question) is to buy from a place that has a tightly controlled supply chain, and look at what may be considered 2nd-tier brands (such as Kingston, PNY, Transcend, etc.).

I have been considering putting a few cameras around my home. I have plenty of reserve storage, RAM and processor power on my server, so I would wish to avoid the cloud and keep everything local. With much of the advertisements, articles and tutorials biased toward cloud solutions or somewhat outdated, I'm reaching out to the Unraid community for ideas and suggestions. What software would you recommend? Docker or VM? Recommendations for camera hardware also appreciated. Need to stay reasonably priced/value for the money. If you are currently running a camera system, what would you do differently? I look forward to everyone's input.

For your initial transfer of data to your array, you will want to: Disable your writes to cache (Primary Storage to array, Secondary Storage to None) for each share you are copying to In Disk Settings, set Tunable (md_write_method) to Reconstruct Write Once you've copied everything over, you can set them to what you wish. Cache will not save you any time if you overflow it, and cause you problems during large transfers. As it will take as long as it takes to write to the array - cache just delays the pain, making it faster in the long term, but then you do the write again later (when you are less concerned about speed). For the Reconstruct write, that does make the transfer faster.

With a i5-11400? A lot. It isn't so much how many VM and docker containers you have installed and/or running. It is more about how hard you are running those activities. A Windows VM sitting idle or browsing the Internet doesn't need a lot of processing power. Plex/Emby streaming a Direct Play video on your home LAN uses very little. Playing a video game in Windows while transcoding several video streams, well that needs much more. My biggest suggestion is to add some more DRAM to your system. 16GB is low if you are running VMs (especially Windows. Would you run Win10/11 on bare metal with 4GB?). 32GB for the system is better. I also added an inexpensive SSD to put my Domains and ISO shares - They run faster from SSD, and it saves space and wear and tear on my primary cache pool.

I would be very happy if those were the only things expected from me. For me, the list keeps on growing. 😆

This is a question that depends not on what you will be running, but how you are running the applications on the server. Running a Win11 VM? Will it just be used to open a few browser tabs and watch a YouTube video, or will you be playing games or doing CPU intensive activities like editing video, etc. Will the Plex be used occasionally to stream Direct Play video for one user on the LAN, or will there be several people watching different videos on devices out on the Internet, all of them transcoding at the same time? Will there be other dockers (such as the arrs) also be active at the time, downloading torrents or Usenet files, while these other activities are taking place? If what you plan to do would only be marginal running on bare metal with an 8th Generation Intel (released 7 years ago), it won't run any better on a server, with several layers of code to execute (Base Linux, Unraid server code, Docker/VM layer code). While it may be a cheaper option, is it really a value to save a few bucks only to decide you need to upgrade to reach the level of satisfaction you desire?

I don't have a solution for your issue, but I do applaud your diligence in testing the hardware. 😄

They probably sent you to this thread - The support thread for Unassigned Devices:

801 Posts and counting here:

First, this is a good thread to review. It is *very* long, but the info in the first post has great info that is still valid today. Both the JBM585 (that I use in my main server) and several of the Asmedia based SATA are all good choices. Note that N5105 boards have a limited number of PCIe lanes, and that many of the boards only use one PCIe lane for the M.2 slot (vs the 4 lanes typically used for M.2 nVME storage). So be sure the SATA adapter you buy matches what the motherboard has electrically.

I've been running a APC BR1350MS for a few years and I have zero complaints. It is plug and play with Unraid and apcupsd. With a fresh battery it reports ~50 minute run time at ~100W load. I recently replaced the battery after 2+ years with the battery for the 1500 model (APCRBC163) - It was a drop in replacement, a few $$ cheaper, and I get some extra A/hr. I don't give too much heed to the UPS reviews I've read. I think that across the board for what is considered consumer UPS, the cheaper models are just that - cheap. Move up the product line some, temper your expectations, and most choices will work out OK.

Sounds like it may only affect you when it needs to transcode. A simple fix that has worked for many is to go into your appdata, tunnel down to the Codec folder and delete all of the files in it. When you start up Plex, it will go out and repopulate these with fresh copies. Many folks have found this to fix intermittent/fluky playback problems. The joys of Plex!

Perhaps because a firewall device, be it a sophisticated data center one, a box running pfsense/OPNsense, or even a cheap home router, is designed to do firewall things. The code is written for routing packets, and those with better code do packet inspection and other bad actor vulnerabilities. These tasks are best run on hardware which is dedicated to this task, both for bandwidth and to reduce possible attack vectors. Unraid is written to be a NAS. It has since gained virtualization and Docker capabilities. With all of this already on its plate (and many folks pushing things to the limit of both the hardware and software), the best advice is leave firewall activities to those focused on writing firewall code. The biggest risk to any system on one's LAN is usually the user who configures and uses the network.

The way that many small companies who distribute OS solve this is to do their damnedest to get users to upgrade to the latest version (Microsoft, Apple, Alphabet,...)

I hope your don't actually have the address as "192,168.x.0/24" - as there is a comma after 192.

Gut feeling it is related to your LSI SAS3008 controller. But I'll let the experts with these see if they agree and offer advice.

contact support, as the message said. This is a community support forum.