FreeNAS stops recognizing me as a user over SMB.

Windows7ge

Contributor
Joined
Sep 26, 2017
Messages
124
You don't touch the top-level dataset's permissions, only the datasets that are being shared.
I'll set it all up and see what happens.

How so? You can pin \\server_hostname and access everything from there. No mappings necessary.
I've always had terrible experiences trying to use the hostname in Network Locations. The server either doesn't show up or half the time when it does it tells me it can't get in with a location not found error. The only thing that works reliably is by IP.

EDIT:
Alright, scratch most of that. The datasets are a sub-directory of the pools root folder so I can just map a drive letter like I usually do. As you said SMB would let me put it all under one share but I can make them individual if I want.
 
Last edited:
Joined
Jan 4, 2014
Messages
1,644
Joined
Jan 4, 2014
Messages
1,644
This has been ongoing for quite a while and it's incredibly annoying. Everything will be working fine for weeks, then for no apparent reason when I login over SMB like any other day it takes my credentials but instead of showing me the root share directory it shows me this:
Has the issue recurred? Have you been able to eliminate the client as the culprit?
 

Windows7ge

Contributor
Joined
Sep 26, 2017
Messages
124
Has the issue recurred? Have you been able to eliminate the client as the culprit?
No. We're going to try reconfiguring the pool permissions, share settings, and invoking datasets which may solve the issue assuming I don't screw with too many permissions again.
 
Joined
Jan 4, 2014
Messages
1,644
Isn't that what we've been discussing having me do? Or are you saying something else...?
@Ericloewe has raised both, but I get the sense there's some ambiguity around the understanding of each. Home directories have a very different meaning https://www.ixsystems.com/documentation/freenas/11.2/accounts.html#users. They exist as directory structures within a dataset. This makes them invisible from the GUI. This then is another disadvantage of home directories, apart from the ones I've already mentioned. You can't, at a glance from the GUI, see how much space is being used by each user. This is a big plus for child datasets (Used column in the example below).

screenshot.143.png
 
Last edited:
Joined
Jan 4, 2014
Messages
1,644
I'll probably keep them all under the same dataset (media + software)
From a conceptual perspective, I'd be inclined to separate software from media. Logically, you can't play software eg. PLEX can't play software, but can play music, movies and pictures. However...

Yeah, but it's going to depend on everyone's use case. Say, family videos versus media - you might want to replicate the former to an extra place that doesn't really justify the expense for the media that could conceivably be re-acquired.
Agreed. I got the feeling @Windows7ge was using the term videos in a generic sense to include movies as well, but yes, separate datasets would allow more selective replication of media type. In my case, I replicate the lot, so it just made more sense to keep all media types under the same media dataset.
 
Last edited:
Joined
Jan 4, 2014
Messages
1,644

Windows7ge

Contributor
Joined
Sep 26, 2017
Messages
124
@Seymour Butt I've rebuilt the pool and ended up creating separate datasets for each software type that I'm storing and I do like how it shows you how much each dataset is using it's a nice feature. I like having the user folders within the viewable GUI because I use it as kind of a catch all, files that don't fit other categories & as a remote access directory that enables me to save them while away from home and I go there when I'm home to retrieve them.

This time I haven't touched the root dataset. It's on default UNIX permissions under root/wheel. I'll set the child datasets to the windows permission type once they're done copying from backup. Hopefully it helps in some way. The only other thing I currently would like to know is how do I setup a login requirement? The videos you linked didn't discuss this.
 
Joined
Jan 4, 2014
Messages
1,644
Joined
Jan 4, 2014
Messages
1,644
It's worth noting that one disadvantage of child datasets over directories is the time that it takes to relocate a file or directory tree - within a dataset, it's almost instantaneous; across datasets, it's noticeably slower.
 

Windows7ge

Contributor
Joined
Sep 26, 2017
Messages
124
It's worth noting that one disadvantage of child datasets over directories is the time that it takes to relocate a file or directory tree - within a dataset, it's almost instantaneous; across datasets, it's noticeably slower.
That won't be too much of an issue for my use case.
 
Top