bfeist

No prob. Attached. telegraf.conf

Thanks for this great tool! I just spent hours working through setting it up. With fresh eyes on the install process I would like to suggest to @falconexeputting version numbers for each docker on the first post (as of today): docker - influxdb:1.8.4-alpine (I see there's an image of this in the first post, but I didn't realize this was a version restriction) docker telegraf - telegraf:1.19.2-alpine Also, including a default telegraf.conf that matches `telegraf:1.19.2-alpine` and the panels in UUD 1.6 would be great. I guess now I need a faster motherboard/CPU too because this dashboard uses ~20% of my resources lol.

Sorry for the delayed reply. Temps are great. Below is an image I just took of current temps, and this is after I added a dust cover across the front of the cases.

Nope. Icy Dock support has no idea what's going on. It's possible the amazon seller who sold me the drive cages was shipping faulty cages.

I bought some foam mesh and wrapped the front to filter dust. Looks pretty good and works great.

The "master" case is more your typical unraid build except I added an external SFF-8088 adapter card that accepted the external cables from the slave and made the connections internally to the SAS card. Still the same old motherboard in the master. Finally, I put all of the drives in the drive trays, connected the two boxes together, and spun up UNRAID. I put the two machines on a cheap set of wheels (I can't believe this thing could take the weight, but it did) https://www.amazon.ca/gp/product/B08R1F9S17/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1 It's tucked away under a desk in the corner, easy to clean under, etc. I'm using negative air pressure in the case for cooling with all air coming in through the drive bays. I plan on building a screen I can place over the front of the two cases to catch dust. I'll post pictures of that when it's done. So that's how you spend a bunch of money on your unraid setup without adding any more drive space! Thanks for reading, and happy compulsive designing. Ben IMG_0493.MOV

The two-case solution would mean that the motherboard was in one case (just as it already was) and I needed to find a way to power and connect the drives in the other "slave" case. I found the solution on the excellent Art of Server YouTube channel. He was selling a SAS expander on eBay that didn't need to be plugged into a motherboard: For the slave case, I bought a PSU (and a simple switch to turn on the PSU without a motherboard) and powered the SAS expander and 3 drive bays. Four SFF-8087 to 4 SATA breakout cables gave me 16 SATA connections in the slave, and two external Mini SAS 26pin (SFF-8088) Male to Mini SAS 26 (SFF-8088) Male cables would allow me to connect the slave to my LSI 9201-16i card, which has 4 SAS SFF-8087 connections. This provides enough bandwidth to the slave for all 15 spinning disks and leaves 8 SATA connections plus the SATA connections on my motherboard (8 of them) for the "master" case drives.

I have no room for a rackmount solution, the cases are too long, even if I converted one to be floor-standing. Plus they're much too loud, even when fans are replaced. I had an extra Cooler Master Centurion 590 case from an old PC--the same model case that I have been using for years for my unraid box. I thought if I could chain the two cases together somehow, I could create a floor-standing tower solution with 30 bays. I bought six Icy Dock MB155SP-B 5 in 3 drive cages--the very devices I had spent all that effort to avoid buying years ago (and wound up spending too much on each unit. Would have been cheaper to just get them years ago). In order to mount these into the Centurion 590s, I had to cut off the 5 1/4 bay tabs with a dremel. I took out all electronics from both cases before doing this and washed the cases down to remove any metal flecks left over to reduce the risk of shorting something out. The finished, slotless 5 1/4 bays had some sharp edges, but nothing that would damage any equpment (only my hands).

I'm a long-time UNRAID user. If i had to guess, I'd say this might be the 7th iteration of my unraid setup. Years ago, I came up with a very inexpensive way to make 5in3 racks I have been using this approach all of this time, slowly replacing my 1tb drives with 2tb, then 4tb, now 8tb (soon 16tb). At the same time I also expanded the number of drives in my mid-tower case to 19. This meant bolting a 4 drive bay with a fan in front of it to the top of my already janky build. It was time to clean things up somehow. My data is precious. Old server case. 124TB with dual parity.

Yes. There are 3 sata power connectors for each unit. They seem to be connected to each other in parallel. I tried powering each of the three connectors individually and a single connector powers the whole unit.

Hi all, I have just added 6, 5in3 MB155SP-B drive cages across two different mid tower cases. They are connected and powered, but the leftmost slot of each of them does not show a green LED and does not power up its drive. I tried swapping SATA data connectors with one of the working slots. I tried powering via a molex to sata convertor. I tried different combinations of drives and connectors. Always no LED, and no power to left drive. I can't imagine that I managed to get 6 lemon drive cages. I must be missing something basic. Does anyone have any suggestions? Thx.

@jondak you're a genius! I just applied the 3.3V fix and they both now spin up. Thanks so much for the quick and knowledgable reply!

Interesting. Wouldn't that have shown up when I first shucked the drive (years ago)? They have both been in my array since ~2019

Hi all, I just moved 19 drives into new ICY DOCK 5 in 3 drive cages. All drives are healthy except two WD drives. One is a WD80EMAZ shucked, the other is a WD80EZAZ shucked. To narrow down the issue, I attached each of them to a USB controller and they spun right up and UNRAID recognized them as part of the array. Putting them back in a drive cage and they wouldn't spin up at all. I have tried different drive cage slots, etc. I have 3 other WD drives in the array that are working. Two are WD80EZZX shucked and one is WD80EZAZ shucked. No issues at all with them. Any thoughts on what might be going on? Thx

I'm a long-time unraid user. I've built several systems over the years for myself. I'm currently running a 18-drive server with custom 5-in-3 trays in a Coolermaster Centurion 590 (plus a cage on top of the case) that sits under a desk. It's time I get a better drive managment system in place with hot-swappable bays and room for more drives. I'm thinking 24-bay would be great, but am open to ideas. I live in a tiny house with no room for a rack-mount system, nor can I afford to have a chassis that's super loud (like most rack-mount chassis are). I was considering putting a second, identical case next to the one I have and using a SAS expander with cables running between the cases. This would give me 30 drive capacity, but this is a step in the "even more janky" direction. I'd prefer to have everything in one case. I've done some searching, but I can't seem to find anything like what I'm looking for. Has anyone seen anything that fits the bill? Thanks!

Great plugin! Chrome randomly reloads the tab clearing the stats history. I'd love it if the stats were kept on the server rather than in my browser so the graphs were always available upon page load. Cheers!

I have the same. Any advice?

Hi everyone, I'm a long-time user (over 10 years at this point) and I'm still using Supermicro AOC-SASLP-MV8 controller cards (two of them) plus my motherboard's onboard SATA connections, of course. I'm in the process of rebuilding a bad parity drive and I'm getting about 65MB/s across 16 drives. Are there newer cards out there that will do better than this? Just wondering if my somewhat ancient hardware is still relevant. Thz

I have the same issue. What is the proper support threat?

After reading all of these terrifying posts about how Promise controllers are deprecated, I decided to try to upgrade anyway. It was completely uneventful, and the 16 promise-based drives I have in the array are none the wiser. Thanks for continuing to push this great software.

Hello. My cache drive is throwing "no space left on device". When I do a df -h, I get: Filesystem Size Used Avail Use% Mounted on /dev/sdb1 239G 85G 0 100% /mnt/cache It's running btrfs. I have done a trim, did a btrfs scrub. No luck. Any thoughts?

This gets even more mysterious. I installed sqlite3 in the linuxserver docker itself and restored the db from dump, then ran the integrity check within the same docker and get a disk I/O error. # sqlite3 com.plexapp.plugins.library.db < dump.sql # sqlite3 com.plexapp.plugins.library.db "PRAGMA integrity_check" Error: disk I/O error ----> update Turns out mapping the folder /mnt/user/appdata/Plex instead of /mnt/cache/appdata/Plex was the issue. No clue why.

Thanks for the suggestion. I completed those steps. Integrity passed. Dumped to .sql then back to .db file. No change. 1.14.1.5488 (needo container) launches Plex fine. 1.15.4.919 (linuxserver container) gives SQLITE3:(nil), 5386, os_unix.c:37072: (19) mmap(/config/Library/Application Support/Plex Media Server/Plug-in Support/Databases/com.plexapp.plugins.library.db-shm) - No such device

Yes, except if I start plex 1.14.1.5488 (needo container) it works perfectly.

New help request. I'm attempting to move from the needo plex docker to linuxserver plex docker. I've pointed the config of the new linuxserver docker to the same /config location as needo. I get a fatal error and nothing will come up. Here's the entire log file when I try to start. Any help would be appreciated. Apr 13, 2019 19:19:54.139 [0x14bf7077e700] INFO - Plex Media Server v1.15.4.919-8e2884e4b - Docker Docker Container (LinuxServer.io) x86_64 - build: linux-x86_64 debian - GMT -04:00 Apr 13, 2019 19:19:54.139 [0x14bf7077e700] INFO - Linux version: 4.18.20-unRAID, language: en-US Apr 13, 2019 19:19:54.139 [0x14bf7077e700] INFO - Processor AMD A10-7850K Radeon R7, 12 Compute Cores 4C+8G Apr 13, 2019 19:19:54.139 [0x14bf7077e700] INFO - /usr/lib/plexmediaserver/Plex Media Server Apr 13, 2019 19:19:54.141 [0x14bf7d583780] ERROR - SQLITE3:(nil), 5386, os_unix.c:37072: (19) mmap(/config/Library/Application Support/Plex Media Server/Plug-in Support/Databases/com.plexapp.plugins.library.db-shm) - No such device Apr 13, 2019 19:19:54.141 [0x14bf7d583780] ERROR - SQLITE3:(nil), 5386, disk I/O error in "PRAGMA cache_size=2000" Apr 13, 2019 19:19:54.142 [0x14bf7d583780] ERROR - Database corruption: sqlite3_statement_backend::prepare: disk I/O error for SQL: PRAGMA cache_size=2000 Apr 13, 2019 19:19:54.142 [0x14bf7d583780] ERROR - Error: Unable to set up server: sqlite3_statement_backend::prepare: disk I/O error for SQL: PRAGMA cache_size=2000 (N4soci10soci_errorE)