I switched to NFS on both my servers, but had to roll back to SMB on my downloads server. While individual and/or smaller files transferred just fine, full seasons or full series took forever just to “prepare”. While this isn’t a huge issue in and of itself, the longer it took to prepare, the more I saw slowdowns in plex. So it was definitely affecting the network or NAS somehow. I believe this might be because I run a VPN on that server, but not entirely sure. But that’s really the only major difference between the two. Maybe I’ll dig into why someday, but the import server to NFS seems to perform well at this point.
Basically Plex would know when a library path is disconnected/unavailable and will not immediately toss it. So for this issue, technically, having Empty Trash enabled shouldn’t remove anything from your libraries. However, I would still recommend caution and disable Empty Trash while observing this issue.
1 Like
Another question to the group. I’m seeing a trend around hibernating drives. Is everyone/most have their NAS set to hibernate/sleep their HDDs? Just wanting to make my test environment match the majority of the affected users here.
Same issue here with no hibernating drives in the NAS.
2 Likes
Perfect! That is honestly all I needed. Just wanted to make sure I wasn’t missing something obvious.
This weekend, I ran a simple Python script to crawl my Movies folder (where I experience this issue) via SMB on my Synology from macOS Sequoia 15.1. It basically just counts the video files and compares it to how many I actually have in that folder. Then, it randomly waits from 30 seconds to 5 hours and tries again. After running all weekend, every query for file counts returned accurate numbers.
I know this experiment doesn’t tell us much, but since I can’t run an actual Plex example (though that would be perfect!), I suspect that this issue isn’t just related to file info requests. It likely involves the way those requests are being made.
1 Like
I believe we’ve confirmed that this issue doesnt happen to just Synology, so I’m not sure if this will help. But it’s definitely worth a shot.
1 Like
2 Likes
That directory leasing part is very interesting. Hope that fixes things…
1 Like
This is my understanding too. However, I believe the majority here are Mac+Synology. If Synology resolves the issue, great! Then we can focus on what others are seeing with other NASs and USB external drives. Sometimes these issues can manifest similarly on differing setups but the root cause be completely different. I’m hopeful for any positive progress from any of the affected components.
1 Like
new Synology SMB update or the latest macOS beta did not resolve the issue.
1 Like
I don’t think it’s a Synology issue. Take Plex and Synology out of the equation, and I’ve been having problems transferring large files (>25GB) from the Mac SSD to an external SSD, but also from the external SSD to the Synology. The transfer fails about 50% of the time, but only when there is other “intense” hdd or network activity happening, like a 4K Remux streaming on Plex, or another simultaneous large transfer happening. I get the unmounted notification, then they mount again quickly (sometimes), but not in time for the xfer not to fail. I never had a problem before 18.0.
1 Like
Hi Folks - First time posting on this issue, but I have been following closely. No tweaks with Synology, Mac OS, or SMB seemed to make a difference for me. However, a week ago I followed the directions on enabling NFS on my Synology and changing out SMB mounts for NFS mounts on my Mac. That was a week ago and so far Plex has worked fine with no issues. I can’t swear this fix will everyone, but it definitely seemed to be the fix for me.
2 Likes
+1 for the NFS workaround. Hopefully a real solution is found, although I am happy with the NFS option so far. Thanks to @alfgut for sharing those instructions. Worked like a charm.
2 Likes
That was my suggestion too… a pre-call to the path with an LS (path) or similar.
Yes. Nice sleeping the drive decreases the MTBF
Not sure if this helps or not. Same issue. Different circumstances.
I am running plex in a docker container on a proxmox server. File storage is on a dell server via truenas in a hyper-v VM with the drives passing through.
My issue started when I migrated plex (via the plex support article) from a windows server running on bare metal to the container above. Everything worked great and the only issue I have had is the above issue (trash can/unavailable) on a single season on 2 different shows. The media plays and the files are present and unduplicated.
Emptying trash doesn’t help.
Reloading the container doesn’t help.
Unmatching/rematching doesn’t help.
I was able to fix it by removing the offending season, scanning the library, replacing the season, and then scanning again.
That is a common issue with Plex (Making you do the “plex shuffle”)
This current issue is related to SMB on macOS 15, where it’s appearing that Plex is not able to get the full file list refreshed before scanning resulting in plex thinking files are missing (trash can or removed if auto delete is enabled).
I don’t personally think it’s related but someone else may be able to find a related link.
Yeah that sounds completely different.
