rutherford

I did NOT know this, thank you very much! I'll get the twins separated. Also comforting to know two docker instances take up pretty much the same amount of space as one.

Wow old thread! But still relevant. Been a long time of this empty photo hosting niche. I’ve been using Immich with Google photos for a few months now. Given the cost of Immich, it blows me away. It’s got machine learning (can recognize faces, beaches, chairs etc), the app is well developed and supported, interface is clean and pretty. if you read the docs over there they recommend the docker compose method, which is a level above my comfort. So I’ve been using the community apps version, and it’s been aok. You do need to connect it to a postgresql docker. I was able to move Nextcloud over to the same postgresql docker as well, so that’s fine. ita a good enough app I’m donating $5/mo to the project via GitHub. I really do believe this is The Way with Immich. its still under lots of development, but in my opinion, it’s good enough for prime time. here’s my thread on getting it working…

@fun.hat I think you're barking up the wrong tree there. Here's the line for Roonserver in my Docker tab I'll slap some more pics down below... Roon > Settings > ARC pic with blue buttons is my Netger Orbi router WAN settings. You gotta make sure your WAN side of your router has an addressable external IP address. Another way you can check this, is visit whatismyip.com from any computer on your network. Then make sure you see that same IP address in the WAN setting in your router. That will make sure that you don't have a double nat situation roon forums. Definitely breeze through some of this stuff, see if anything jumps out at you as applying to your situation https://community.roonlabs.com/c/support/port-forwarding-resources/104 And finally the port forwarding services that I have is the one with all the redactions. My server ip is 192.168.11.54 oh, and the docker settings, with More Settings (?) expanded at the bottom. Something else you wanna watch out for is cgnat: carrier grade NAT, that means the IP address you've been given IS NOT accessible from the www. Check out lists of ISPs to see if you are in that situation. Might be that a static IP from your ISP would sort that out. Or some tailscale network wide VPN would also sort that out.

Got a few really neat questions from someone via craigslist or FB market place or something. I wanted to introduce them to our Extra community here: one of the biggest value-adds unRaid has to offer. ============================================================================== You must be quite the audiophile . . . I was looking at your listings on eBay and those are high end items! And completely positive feedback from buyers -- very nice. I came across an old box I already have which might accommodate the bare UnRAID minimum, which I believe would be cache/parity/data drives. If that doesn't accommodate those drives, I may well get back to you about your box. Obviously it could sell at any time. I'm hoping that UnRAID becomes less imposing as one gets into it. And I'll take your advice re: joining the UnRAID forum for help with questions there. One last question: did you use the piKVM for controlling your UnRAID box from afar? or for other purposes? Some of the YouTube videos I've seen make piKVM sound very, very cool. ============================================================================== Howdy! I do fancy myself an audiophile yeah. I love those high quality FLAC tracks, some vinyl and spending too much money on stereo equipment. I figure it could be worse... ¯\_(ツ)_/¯ Love me some flattery as well of course. I think the best way to get started here with unRaid is the easiest route: just put together a computer with more than one hard drive, boot off the flash drive, and use it as Network Attached Storage for now. At your own pace you can start customizing drive pools, SMB (samba windows) sharing stuff, maybe a few other things. Unraid connect forum post detailing that piece Then move onto Docker and the Community Apps store. That's where things will start to get more difficult, and cool! Spacedinvader1 is a fabulous resource. Some videos are a bit dated, but still work great. This looks like a great place to start: I'll slap more info in here as I think of it. Welcome to the community! ruth

thanks for getting back JorgeB. Ran that, didn't see any smoking guns:

Had a the server hard crash, and I just wanted to make sure I was on the right track here: https://forums.unraid.net/bug-reports/stable-releases/691-metadata-corruption-detected-at-xfs_dinode_verify-r1369/ just wanna make sure I'm on the right track here. thanks! https://wiki.unraid.net/Check_Disk_Filesystems#Checking_and_fixing_drives_in_the_webGui - Settings -> Docker -> Host access to custom networks: Disabled - Use custom network "proxynet" for ALL docker instances (before: host + bridge + proxynet) - might be unrelated mayorgoodway-diagnostics-20240127-1116.zip

Perfect, thanks! Found that over at the tiredofit github docs Added this VARIABLE: DEFAULT_FILESYSTEM_PATH_PERMISSION, as 0755, that should do the trick! thanks again

db-backup. The files it creates are -rw------- 1 10000 10000 19203997 Jan 15 20:24 pgsql_nextclouddb_192.168.11.54_20240115-202405.sql.gz owner: rw, group: nothing, everyone: nothing how would I change this so a different docker can read the files and copy them to an external backup drive? chmod 777 those files?

Another crash this morning: "General Protection Fault." I'm thinking it's the RAM. Searched for that error and that user had a RAM issue. Various logs attached. I'll swap the RAM out just in case. mayorgoodway-diagnostics-20240106-2117.zip mayorgoodway-syslog-20240107-0516.zip mayorgoodway-syslog-previous-20240107-0516.zip

bump

Maybe some of this changed now with the later version? unRaid 6.12.6 1/2024 - user from the future!!!

moving install from one server to another. I'm adapting the instructions at https://immich.app/docs/administration/backup-and-restore command pg_dumpall dumps ALL the databases. I don't want to restore ALL databases, because I'm also using the postgresql15 docker for my nextcloud install. I imagine that the reason we're also dumping the postgres database is that it contains user information. I'll have to manually re-add the database user before proceeding. https://www.postgresqltutorial.com/postgresql-administration/postgresql-backup-database/ Another funny thing: Immich makes a point of NOT modifying my original files. Generally I appreciate that, but when I got my junk out of Google Takeout they were missing all sorts of stuff! I had to jump through several hoops to make the import meaningful with the missing metadata. If that process could just make those metadata changes CORRECT, and make changes to my files, the import, and future migrations, would also work smoothly. Get your database name from existing Immich docker>edit. Mine is immichdb Console into your postgresql15 dock For ease of getting the backup file out of the docker, I'm going to move over to the appdata connected folder at /var/lib/postgresql/data, which is mapped to /mnt/cache/appdata/postgresql15 on my unRaid server. Also by the by: you don't connect to your database server to run this command, only the console of the docker. # cd /var/lib/postgresql/data # pg_dump -U postgres -F t immichdb > immichdb.sql.tar # ls -lh | grep immich -rw-r--r-- 1 root root 807M Jan 6 08:53 immichdb.sql.tar I'm going to eat my shorts if this works.... and this is why I have doubts: "Note that for the database restore to proceed properly, it requires a completely fresh install (i.e. the Immich server has never run since creating the Docker containers). If the Immich app has run, Postgres conflicts may be encountered upon database restoration (relation already exists, violated foreign key constraints, multiple primary keys, etc.)." location of jpg file library oldserver:/mnt/user/memories/immich/library/d8435a7b-c642-46c7-9f84-2f5832215e48# rsync'ed to newserver:/mnt/user/memories/immich/library/d8435a7b-c642-46c7-9f84-2f5832215e48# (I have a hunch on the new install that file location d8435a7b-c642-46c7-9f84-2f5832215e48, will be a different random string... I'll see about how to get at that later...) oldserver immich docker Path /photos > /mnt/user/memories/immich Get into the postgresql docker CLI # psql -U postgres -W Password: psql (15.5 (Debian 15.5-1.pgdg120+1)) Type "help" for help. postgres=# note: postgresql15 doesn't require a password when connecting from localhost <shrug> postgres=# \list List of databases Name | Owner | Encoding | Collate | Ctype | ICU Locale | Locale Provider | Access privileges -------------+-----------+----------+------------+------------+------------+-----------------+------------------------- immichdb | immich | UTF8 | en_US.utf8 | en_US.utf8 | | libc | =Tc/immich + | | | | | | | immich=CTc/immich nextclouddb | nextcloud | UTF8 | en_US.utf8 | en_US.utf8 | | libc | nextcloud=CTc/nextcloud postgres | postgres | UTF8 | en_US.utf8 | en_US.utf8 | | libc | template0 | postgres | UTF8 | en_US.utf8 | en_US.utf8 | | libc | =c/postgres + | | | | | | | postgres=CTc/postgres template1 | postgres | UTF8 | en_US.utf8 | en_US.utf8 | | libc | =c/postgres + | | | | | | | postgres=CTc/postgres (5 rows) There's my immichdb, nuke it. postgres=# DROP DATABASE immichdb; DROP DATABASE postgres=# \list List of databases Name | Owner | Encoding | Collate | Ctype | ICU Locale | Locale Provider | Access privileges -------------+-----------+----------+------------+------------+------------+-----------------+------------------------- nextclouddb | nextcloud | UTF8 | en_US.utf8 | en_US.utf8 | | libc | nextcloud=CTc/nextcloud postgres | postgres | UTF8 | en_US.utf8 | en_US.utf8 | | libc | template0 | postgres | UTF8 | en_US.utf8 | en_US.utf8 | | libc | =c/postgres + | | | | | | | postgres=CTc/postgres template1 | postgres | UTF8 | en_US.utf8 | en_US.utf8 | | libc | =c/postgres + | | | | | | | postgres=CTc/postgres (4 rows) postgres=# cool. Now import other one... I tried to rsync a single file, but ended up just dragging from oldserver to Windows desktop, then up to newserver. # psql -U postgres -W Password: psql (15.5 (Debian 15.5-1.pgdg120+1)) Type "help" for help. postgres=# CREATE USER immichuser WITH PASSWORD 'immichpass'; CREATE ROLE postgres=# CREATE DATABASE immichdb TEMPLATE template0 ENCODING 'UNICODE'; CREATE DATABASE postgres=# ALTER DATABASE immichdb OWNER TO immichuser; ALTER DATABASE postgres=# GRANT ALL PRIVILEGES ON DATABASE immichdb TO immichuser; GRANT postgres=# ALTER USER immichuser WITH SUPERUSER; ALTER ROLE postgres=# pg_restore -c -U postgres -d immichdb -v immichdb.sql.tar -W and by god it worked. I can't believe it. I really don't recommend doing this. What a PITA. buuuuut <shrug> seems to have worked! My settings: also /mnt/user/appdata/swag/nginx/proxy-confs/immich.subdomain.conf only changes below are server_name. I thought I'd have to change upstream_port, but nope! upstream_app already matched docker name.

So here's the thing I have going on now: moving this install to a new system. When I did my Google Takeout, it was a shitshow. Awful dates and metadata. Once I ran it through the process there, it was way better. Immich has this technique of saving all it's metadata separate from the original files. Good when you don't want to mess with those, but bad when you're migrating. You must also move the database over. There are decent instructions for it, but they really are tailored towards the official install method, which I've not chose here: I went with our Community Applications template, imagegenius docker. https://immich.app/docs/administration/backup-and-restore My previous installation was postgres15, as is this one, and they are both up and running. But: "Note that for the database restore to proceed properly, it requires a completely fresh install (i.e. the Immich server has never run since creating the Docker containers)." I'm going to have to nuke this new one, and try again. Anyone already do this procedure and can share some steps, with relation to the CA imagegenius version? thanks!

Just did a full server replace, and this server is ready to run as is. I pulled the GPU, and the piKVM interface. "rubble" has been a solid server for years, but it is old. I'm doing 50% off retail price for components. This thing weighs 50lbs, and I don't have a chassis box! Pickup is preferred. I'm going to list this locally too. I've included a bunch of pics in the google album below, and a few direct attachments to this post. chassis Antec P380 mobo Asus ROG STRIX B350-F GAMING memory 2x16GB "fury" ram cache Samsung 980 nvme 1TB CPU AMD Ryzen 7 1700 Eight-core @ 3000 MHz PSU Corsair RM750X controller (might be a) SUPERMICRO AOC-SASLP-MV8 PCI-Express x4 Low Profile SAS RAID Controller. I'm not 100% sure on what this is. hard drives 6 data, 1 parity, and 1 unopened new extra drive. 8TB seagate drives, see pics. Again with these, I'm 90% sure they all match up. The most recent pic in the album is also attached to file name IMG_6283.JPG is the server as it is, and will be off to you. I still need to wipe all the drives. My username at eBay is dkerlee, you can see I'm a long time contributor here on the unRaid forums. You're order and cash is safe with me. I am willing to part this out. Prices will be a little higher than the full-meal-deal. https://photos.app.goo.gl/GgKzSktsbW5gie6i6 Samsung_SSD_980_1TB_S64ANG0R334057L-20231230-1257.txt ST8000DM004-2CX188_WCT01MFN-20231230-1255.txt ST8000DM004-2CX188_WCT02JGR-20231230-1255.txt ST8000DM004-2CX188_WCT02XA2-20231230-1255.txt ST8000DM004-2CX188_WCT03HHD-20231230-1254.txt ST8000DM004-2CX188_WCT025VR-20231230-1255.txt ST8000DM004-2CX188_ZCT0WJY1-20231230-1256.txt ST8000DM004-2CX188_ZR11LJCF-20231230-1256.txt

I have a couple of questions: 1. enable UEFI boot method, how do I know if I need to do this? 2. the "manual method" referenced above has us download the original unraid boot files from the website: what about all our settings, disk allocations, all that stuff? I thought even the manual method would involve a backup Flash from the existing server. derp, this is probably that step: https://docs.unraid.net/unraid-os/manual/changing-the-flash-device/#manual-method "Before removing the flash from the PC, copy the boot/config folder from the backup you made, into the flash drive, overwriting existing files." to possibly answer my #1 question, I can look and see how I'm currently booting under Main > Flash, go to bottom for this. Server boot mode: Legacy Permit UEFI boot mode <checkbox> mine is unchecked.

a post for posterity. I'm on a new install of unraid (6.12.6) at the tail end of a total server migration. The Unassigned Devices and Unassigned Devices + both showed as installed in Apps tab. There was nothing available in Main, or in Settings > Unassigned Devices. Not a thing. Looked like below. Fix: also under Apps, UD and UD+ showed updates available. When clicked Actions > Update, it went through the PLG update, and badda bing, everything started working. Just a little oddity. <shrug>

I'm also curious about this. I'm moving my nextcloud install to a new server (Nextcloud Hub 7 (28.0.1) ). I'm also changing the database connection over from mariadb/mysql to postgresql15. I decided not to bother with the database migration, and am just going to rebuild. I have three users: admin, ruth, and ann. I figured I'd create the users, then point the entire storage directory over at my /mnt/user/nextcloud share. Not sure what the hell will happen with permissions there - I need the ruth files to be there for ruth, ann files for ann, etc. <throws hands in air> Let's mess around and find out!

done. thank you

actually everything has changed. I'm migrating all my dockers, settings and files to a new server altogether. The CPU and motherboard is slightly not as awesome as my previous, but the build is way nicer (more experience!), and the drives are 14TB instead of the 8TBs I had on the older machine. RAM: thank you. That gives me something to look at. Holy smokes is 99% function RAM a PITA. Rule for me: never skimp on RAM.

well that was unexpected, had a hard crash this morning, and it wouldn't respond to anything but pulling the power. I am planning on swapping out (slight upgrade) the motherboard and CPU, may as well get on that today while it's down. Any ideas on what caused this crash? I could focus on memory or the cpu etc.? The hardware is gonna get sold, and I also need to have a caveat with it if I'm selling iffy hardware!

Just wanted to post the steps I followed to get nextcloud, SWAG, and postgresql15 all working. Installed this nextcloud docker with default settings and ran web setup one time. Got into postgresql15 console psql -U postgres -W CREATE USER nextcloud WITH PASSWORD 'nextpass'; CREATE DATABASE nextclouddb TEMPLATE template0 ENCODING 'UNICODE'; ALTER DATABASE nextclouddb OWNER TO nextcloud; GRANT ALL PRIVILEGES ON DATABASE nextclouddb TO nextcloud; Then went back to swag subdomain config file, where it suggested I add in some info to /mnt/user/appdata/nextcloud/www/nextcloud/config/config.php <?php $CONFIG = array ( 'datadirectory' => '/data', 'instanceid' => 'oc0w4lt90dch', 'passwordsalt' => 'xxxxxxxxxxxxxx', 'secret' => 'xxxxxxxxxxxxxxx', 'trusted_domains' => array ( 0 => '192.168.11.54:444', 1 => 'nextcloud.mydomain.com', ), 'dbtype' => 'pgsql', 'version' => '28.0.1.1', 'overwrite.cli.url' => 'https://nextcloud.mydomain.com/', 'dbname' => 'nextclouddb', 'dbhost' => '192.168.11.54:5432', 'dbport' => '', 'dbtableprefix' => 'oc_', 'dbuser' => 'nextcloud', 'dbpassword' => 'nextpass', 'installed' => true, 'trusted_proxies' => array ( 0 => 'swag', ), 'overwritehost' => 'nextcloud.mydomain.com', 'overwriteprotocol' => 'https', 'memcache.local' => '\\OC\\Memcache\\APCu', 'filelocking.enabled' => true, 'memcache.locking' => '\\OC\\Memcache\\APCu', 'upgrade.disable-web' => true, ); added the subdomain to my CNAME in DNS hosting, added the "nextcloud" to the SWAG config, restarted, bob's your uncle. Once I logged into the admin panel > Overview, there are many things to fix in there, I'll start adjusting those settings like the suggestion next.

Is this in lieu of installing the other docker container? would you post some screen shots of what the unraid docker settings look like? Thanks!

Migrated to a new server this evening. It did not go well. I ended up with a fresh install. Previous install was working, so I was able, via the windows chrome plugin, to Settings > Export Vault to json file. Re setup the users, login, and restore from the json file. Works, but wonk-city. This guide was useless: https://bitwarden.com/help/migration/#tab-host-to-host-52Q80N79LCU2kkRJpuPKMy I couldn't find the bwdata/ anywhere, or the uid whatever. <bah> it's done.

Anyone have a fix for the new "vector" requirements in the database connection? Besides the different repository? Or is that the best/only way to do it.