Started By
Message

re: more microsoft hell - windows 10 vm edition

Posted on 10/20/26 at 8:25 am to
Posted by CAD703X
Liberty Island
Member since Jul 2008
95346 posts
Posted on 10/20/26 at 8:25 am to
quote:

massgrave is definitely better for CAD

Thanks baw. Registered and enrolled in esu. That was easy.
Posted by CAD703X
Liberty Island
Member since Jul 2008
95346 posts
Posted on 10/20/26 at 8:32 am to
quote:

One has zero access to adult VOD, other can only see adult VOD, makes libraries clean.
no pun intended right?

I hope you named that service "90s video rental store back curtain area"
Posted by bluebarracuda
Member since Oct 2011
19614 posts
Posted on 10/20/26 at 10:18 am to
What handles your network shares? If your disks live in the server (even better if they are connected to an HBA), pass em through to a TrueNAS VM
This post was edited on 10/20/26 at 10:19 am
Posted by Dallaswho
Texas
Member since Dec 2023
3883 posts
Posted on 10/20/26 at 11:18 am to
I haven’t seen that PVE screen in 18 months. Gives me n the creeps.

I’m sure it’s fine. A lot of people use it but I don’t think it makes sense for me as I run 2x very different core ultra minis and at least 12 services using media or AI acceleration or both. Plus I have all NVMe storage and only 3TB live. I back up app data and that’s it.

Has anyone figured out how to make radarr hard link media on NFS shares instead of copying it? Both the qbit category storage and the media folder live on the NFS mount. Simple enough to just put a watchdog on the host but that seems wrong.
This post was edited on 10/20/26 at 11:18 am
Posted by CAD703X
Liberty Island
Member since Jul 2008
95346 posts
Posted on 10/20/26 at 1:11 pm to
quote:

What handles your network shares? If your disks live in the server (even better if they are connected to an HBA), pass em through to a TrueNAS VM
shitty external USB spin drives but I have a terramaster nas I can't afford to fill up right now that will be raid6 I'm moving everything over to.
Posted by CAD703X
Liberty Island
Member since Jul 2008
95346 posts
Posted on 10/20/26 at 1:13 pm to
quote:

anyone figured out how to make radarr hard link media on NFS shares instead of copying it? Both the qbit category storage and the media folder live on the NFS mount.
I'm not sure I follow but my shares follow a movies, TV music simple configuration and qbit writes directly to them from the arrs and the media shows up in Plex a few minutes later.

i dont pull things into one place then move it/copy/whatever. it comes in to where it is served from.

eta in short:

/movies <- qbit, plex, prowlarr, radarr seerr
/tv <- qbit, plex, prowlarr, sonarr seerr
/music <-qbit, plex, prowlarr, lidarr

This post was edited on 10/20/26 at 1:48 pm
Posted by Dallaswho
Texas
Member since Dec 2023
3883 posts
Posted on 10/20/26 at 1:45 pm to
something like qbit to /mnt/nfs/share/downloads/raqdarr1 then once finished, radarr should hard link with possibly cleaner file name to /mnt/nfs/media/radarr1. That way qbit still seed and media server gets a clean grab. I think that's the ""standard"" workflow. whatever that means. im not really into this stuff. It works on 1x radarr that is a "local" as in docker binds to local / but it is copying instead of hard linking on the one that binds into /mnt/nfs/... the link happens on the NFS host so maybe it is an NFS issue not getting a link command or something, not sure.
Posted by bluebarracuda
Member since Oct 2011
19614 posts
Posted on 10/20/26 at 1:47 pm to
quote:

raid6


Stinky. Stick with zfs
Posted by CAD703X
Liberty Island
Member since Jul 2008
95346 posts
Posted on 10/20/26 at 1:50 pm to
quote:

Stinky. Stick with zfs


proxmox system/zfs/raid1

nas spin drive/raid6

everything i read about how drives purchased at the same time tend to fail based on their 'born on' date makes me want that second level of redundancy while the 'failover drive' is being thrashed to death restoring the one that went down for a solid 2 days because if that one fails..then everything is gone. call me paranoid but i've been recently burned by drives that were fried during the ice storm here in january. sure i lose a ton of storage space but i've steeled myself to only getting 40TB of use out of 4 20TB drives.
This post was edited on 10/20/26 at 2:01 pm
Posted by CAD703X
Liberty Island
Member since Jul 2008
95346 posts
Posted on 10/20/26 at 1:57 pm to
quote:

something like qbit to /mnt/nfs/share/downloads/raqdarr1 then once finished, radarr should hard link with possibly cleaner file name to /mnt/nfs/media/radarr1. That way qbit still seed and media server gets a clean grab. I think that's the ""standard"" workflow. whatever that means. im not really into this stuff. It works on 1x radarr that is a "local" as in docker binds to local / but it is copying instead of hard linking on the one that binds into /mnt/nfs/... the link happens on the NFS host so maybe it is an NFS issue not getting a link command or something, not sure.
long time plex user here short time understanding how any of this crap works.

but imo, i dont care what the files are named. once upon a time i tried to fix file names then decided plex could almost always figure it out and it was just easier to make the fixes in the UX for oddballs.

that works much better for movies/tv than music. i still get a '10 cd compilation' and end up with shite everywhere and i have to 'collect' then merge everything to get it into a single title in plex and the real fun..it thinks all 10 are cd1 so i get to manually go in and find the correct cd for each track and change the CD field in the UX.

maybe lidarr could fix some of that with renaming but thats all new to me. been pretty happy over the last 2 decades letting plex do most of that for me and not giving a damn what the filenames are and don't have ot worry about qbit losing the torrent reference.

protip: its stupid easy to get ratio on IPT so set your system to delete it the second it hits 1.0 ratio. or if 2 weeks have passed, do a mass "delete torrents' in qbit. no need to seed anything from IPT after that; you won't get a 'hit and run' message that way for any torrents that haven't reached 1.0 ratio.
This post was edited on 10/20/26 at 2:04 pm
Posted by Dallaswho
Texas
Member since Dec 2023
3883 posts
Posted on 10/20/26 at 2:20 pm to
I'm finding it extremely hard to get ratio.. but I'm behind VPN so that probably cuts out 75% of the people I can connect with because they can't connect to me, i must connect to them. I have hit and run protection so oh well, i mainly get free stuff and if it's not free, i go small.

good point, i can probably just point everything to the qbit folder and get rid of renaming and then it wont link or copy.


Posted by CAD703X
Liberty Island
Member since Jul 2008
95346 posts
Posted on 10/20/26 at 2:29 pm to
quote:

I have hit and run protection so oh well, i mainly get free stuff and if it's not free, i go small.

if you pay $10 for their IPTV service that takes care of that; or maybe thats what you're saying

and yes, almost everything on IPT has multiple freeleech versions. just hit those for now and one day you'll notice your ratio in IPT has magically worked its way up to 4 or 5 and you never have to worry about it again.

now music sites..fricking hell those are hard to get ratio from.
Posted by Dallaswho
Texas
Member since Dec 2023
3883 posts
Posted on 10/20/26 at 3:05 pm to
yup. vip and radarr set up with "free" profile that requires custom format "free" that requires that tag. seer only requests free profile.

so i try to point everything back to /downloads and i hit another snag... some are just files not folders and radarr only allows one folder to one movie ratio... more hit and run time.

edit: ok no hit and run... leaving qbit as source of truth to extent possible... so just remove non compatibles from radarr, but i have some stranded in a /media due to a previous blunder.

I freed 350g of redundant files. Thats huge for my ssd only stack. And everything works correctly now. Time for some more 4k remux for the brand new TV. Surprised my little $30 onn 4k+ is actually handling 100+ mbps video. Or at least it seems fine. The VLC graph shows it keeping up and it looks good to me. Making me wonder what people are doing with their Apple TVs and shields and what not.

Edit again: original problem was 99% because I did a /media docker bind and a /downloads so the container saw different file systems. If I did one bind for /mnr/nfs/ instead then it probably would have worked. I don’t know what that’s 100% clear to me when sitting on the couch, but can’t figure it out while in the computer.
This post was edited on 10/20/26 at 9:43 pm
Posted by CAD703X
Liberty Island
Member since Jul 2008
95346 posts
Posted on 10/20/26 at 7:49 pm to
That little onn box is killing it for me too
Posted by bluebarracuda
Member since Oct 2011
19614 posts
Posted on 10/20/26 at 7:58 pm to
My ratio has gone from ~5 to ~2 over the years, but I still have about 9tb of headroom before I get to a 1 ratio
Posted by Dallaswho
Texas
Member since Dec 2023
3883 posts
Posted on 10/20/26 at 8:10 pm to
Ha. I’m near a 3.0 but have 100g headroom.
Posted by CAD703X
Liberty Island
Member since Jul 2008
95346 posts
Posted on 10/20/26 at 8:16 pm to
I just had another brainstorm with my huge home network project. Because we have everything from hardware to vms to smart doodads and they all demand attention:

- what am I?
- am I running properly?
- what is my version?
- am I ready for an update?
- where do I live?
- how am I protected? Backup? Redundancy?
- who depends on me?
- who gets notified if I have a problem?


So I'm going to take everything and turn it into an API

Everything gets to answer those basic questions and with that we can populate a dashboard or webpage or whatever.

This json response will be the source of truth and the only way to preserve my sanity after everything is live. Everything uses the API.

Still in development but it's going to be sweet when it's done.
Posted by bluebarracuda
Member since Oct 2011
19614 posts
Posted on 10/21/26 at 8:12 am to
Probably already have applications built to do those (Uptime Kuma, Prometheus + Grafana, Zabbix) that can be integrated into HA dashboards. I had Claude build out a tool called Switchdex that handles a lot of this for your networking infrastructure
Posted by CAD703X
Liberty Island
Member since Jul 2008
95346 posts
Posted on 10/21/26 at 8:40 pm to
quote:

Probably already have applications built to do those (Uptime Kuma, Prometheus + Grafana, Zabbix) that can be integrated into HA dashboards. I had Claude build out a tool called Switchdex that handles a lot of this for your networking infrastructure
i'm barely using HA anymore except to pull in the entities and send them to custom html pages.

so you tihnk that is best; here is a rough idea of the stuff
* appliances (all thermador one connect + wifi ge washer/dryer) Not my house
* tplink router, controller, 4 switches, 3 access points
* zwave+ controller, 4 zwave+ thermostats, 4 schalge locks, leak detector & valve shutoff device
* all the HA integrations, OS
* proxmox and all its nonsense, vms, tecnitium, tailscale, plex, arrs out the arse, qbit
* 10-12 reolink cameras & nvr


tons of shite i'm leaving out but there's a signiicant amount of shite to keep updated and happy
Posted by Dallaswho
Texas
Member since Dec 2023
3883 posts
Posted on 10/21/26 at 8:51 pm to
quote:

i'm barely using HA anymore except to pull in the entities and send them to custom html pages.


my man.

frick home assistant. evil necessity. run it but work from outside in.

also frick infra. run all debian and docker. back up app data offline. no infra other than a couple docker nets.

really hope i don't get attached to me media or this wont work.
This post was edited on 10/21/26 at 8:54 pm
first pageprev pagePage 2 of 3Next pagelast page

Back to top
logoFollow TigerDroppings for LSU Football News
Follow us on X, Facebook and Instagram to get the latest updates on LSU Football and Recruiting.

Facebook•X•Instagram