Here is the same problem. Mac Mini M1 with a Synology Diskstation.
We are not just a few 
5 Likes
Meanwhile, back at the ranch, Tammy Trueheart… oops wrong digression.
…
Meanwhile^2: Leaving 2 folders open (Movies and TeeVee) has continued to work as a stop gap for the 3 days since last reporting on it.
Conclusion… (possibly) a timeout issue that doesn’t wait long enough after sending a request to a remote drive. By the time the drive spins up, Plex has stopped waiting and errored out.
2 Likes
Re: Replicating the event
I’ll buy pizza if you folk want to come and look at my server FtF. ;^)
1 Like
THANK YOU!
1 Like
It seems Time Machine backups over SMB to NAS are affected as well and some people have reported disabling the in-built firewall helps.
3 Likes
So this was going to be part of my next survey to the group.
Are you also running Time Machine backups to the same mount location (NAS or USB) where you’re media is stored?
Another ask from me, in the case of users with SMB shares (NAS) could you try something for me?
- Open and start logging in the Console app in MacOS and set a search for
smb - Walk through your steps to reproduce the issue and monitor the Console output to see if any errors or messages point to an issue on the smb mount.
The testing environments I have setup does have some limitations that make it difficult to match 1:1 what real users would be running. The other missing pieces here is all the other real user use cases like Time Machine or any other services that users would also be running along with Plex.
Welp, just added some movie rips and scanned, library back down to under 1000…Finder windows open, NAS folder mounted and available.
1 Like
So I am funning TM to a seperated shared folder/mounted volumes on my Synology NAS. Backups have been going fine.
I am seeing SMB issues. Oddly, Plex is triggering a TM issue, but TM is not backing up.
error 11:51:32.195409-0500 kernel smbfs_vnop_ioctl: Time Machine entitlement check failed
I also see this a lot:
error 11:48:50.140825-0500 kernel smb2fs_smb_cmpd_query: smb2fs_smb_ntcreatex failed 13
error 11:48:50.140830-0500 kernel smb2fs_smb_cmpd_create: smb2fs_smb_ntcreatex failed 13
I am more than happy to send a txt along with the smb issues.
4 Likes
here are the console logs with SMB showing issue with finder window closed, and then reopened and scanned again to locate missing items.
I use: Connectmenow4 to keep mounts alive (closed/quit for these tests)
Time Machine : made sure no backup was in progress for these tests
Athena = NAS
Core = Media volume
Archive.zip (3.2 KB)
Good Scan.rtf = logs for SMB from console with Finder windows open (all items found)
Bad Scan.rtf = logs for SMB from console with Finder windows closed (missing items)
to add, this Mac basically only runs Plex.
4 Likes
I have Time Machine configured on my Plex server, but backing up to a locally connected USB drive.
2 Likes
Same here. Backing up to a different external. Still have had the issue since the beta before the full release.
1 Like
I’m not a smart man, so I have no idea if this is helpful. Here is what I see when I attempt a scan after all the trash can men appear.
error 21:13:08.483190-0700 kernel smbfs_vnop_ioctl: TM Read - attributes 0x1c reconnTO 0 QoS 0x0
error 21:13:08.485404-0700 kernel smbfs_vnop_ioctl: TM Read - attributes 0x1c reconnTO 0 QoS 0x0
error 21:13:09.032216-0700 kernel smbfs_vnop_ioctl: TM Read - attributes 0x1c reconnTO 0 QoS 0x0
error 21:13:09.148041-0700 kernel smbfs_vnop_ioctl: TM Read - attributes 0x1c reconnTO 0 QoS 0x0
error 21:13:09.285671-0700 kernel smbfs_vnop_ioctl: TM Read - attributes 0x1c reconnTO 0 QoS 0x0
error 21:13:09.286451-0700 kernel smbfs_vnop_ioctl: TM Read - attributes 0x1c reconnTO 0 QoS 0x0
default 21:13:17.229119-0700 NetAuthSysAgent URL = smb://SynologyNAS._smb._tcp.local
default 21:13:17.275024-0700 NetAuthSysAgent GetServerInfo serverParamsDict = {
GuestOnly = 0;
MechTypesSupported = {
SPNEGONegTokenInitHintsHostname = “not_defined_in_RFC4178@please_ignore”;
SPNEGONegTokenInitMechs = {
“1.3.6.1.4.1.311.2.2.10” = {length = 0, bytes = 0x};
};
};
MountedByUser =
MountedWithAuthenticationInfo = 1;
NetFSMachineType = Xserve;
SMBNativeOS = UNIX;
ServerDisplayName = SynologyNAS;
SupportsChangePassword = 0;
SupportsGuest = 1;
}
Here is what I see when I try the open finder “workaround”.
default 21:13:17.301453-0700 NetAuthSysAgent Found Server Marker for “SynologyNAS._smb._tcp.local”
default 21:13:17.319080-0700 NetAuthSysAgent MechType session created for host “SynologyNAS._smb._tcp.local”, service “cifs”.
default 21:13:17.319971-0700 NetAuthSysAgent server smb://SynologyNAS._smb._tcp.local
error 21:13:17.362340-0700 NetAuthSysAgent already_mounted: share <> already mounted
error 21:13:17.362402-0700 NetAuthSysAgent already_mounted: share <> already mounted
I empty the trash, then I try to scan again, and I get 20+ lines of this again.
error 21:30:15.853115-0700 kernel smbfs_vnop_ioctl: TM Read - attributes 0x1c reconnTO 0 QoS 0x0
Something I didn’t see mentioned - the trash can men appear in the Plex App for MacOS. But when the mouse pointer is hovered over the movie, the trash can disappears, and you can select play and everything works. On Apple TV, the problem is the opposite - no trash can man (or “unavailable” tag), but Plex says the file does not exist.
3 Likes
I noticed during the rebuilding/troublshooting phase that my backups were not current. I redefined them to a local path for just that reason.
…
Another date point. I reported earlier that I was seeing the mapped remote paths reverting back to the root (or a local drive) on my work machine (Mac Mini Intel)…
I started making routine OS calls for different housekeeping events (free memory, free purgable space, etc.). What repeatedly causes the reaction was a call to flush the DNS cache.
Given we know what might be a time out, and this, perhaps a pre-call before each drive access is needed?
If the developers are making calls to libraries or OS functions that they can’t directly control, they can act before sending the call to the “black box”.
- Do a confirmation that the mapping as defined in setting remains (remap, etc.)
- Do a pre-call to the content listing (LS [path]) to wake up sleeping drive(s).
Might it also address a router cache flush that reassigns addresses? I would think named paths and fixed IPs are more common. It might explain why the mapping acts this way when flushing the DNS cache on a machine.
…
“Something I didn’t see mentioned - the trash can men appear in the Plex App for MacOS. But when the mouse pointer is hovered over the movie, the trash can disappears, and you can select play and everything works. On Apple TV, the problem is the opposite - no trash can man (or “unavailable” tag), but Plex says the file does not exist.”
The thread’s getting really long, but I too reported this trait. I also reported that a file that can’t be played on a remote client can be started on the host server (the TMC changing to Play), FF to some part about a 1/4th in, then it will appear in “continue watching” and play on all remote clients.
2 Likes
BTW: turning off Time Machine does not resolve the issue.
3 Likes
I’m not using time machine at all. Same issues as everyone else.
3 Likes
And what about your firewall?
1 Like
MacOS Firewall disabled.
2 Likes
Mac firewall is disabled by default.
2 Likes
It’s been 3 days since I updated to MacOS 15.0, and I experienced this problem immediately. I thought changing to NFS was the solution, since I haven’t had the problem for a few days, but it’s not. The common denominator solution is keeping a finder window open. But if the finder window is closed, and a scan happens, the library gets borked, and reopening the finder window and rescanning does not recover it, even if emptying the trash before rescanning. I have to edit the library in Plex and add the same path again, then the rescan is successful.
Also, it only seems to larger libraries. What that # is - I don’t know, but I have a 4K Remux library with 1K movies and a compressed library with 7K movies. The 4K library was unaffected. Same story with music. One library with 1K albums, and another with 9K albums - only the larger library was affected.
2 Likes
Ensure you have “Empty trash automatically after every scan” disabled in Settings → [Server Name] → Library (Show Advanced). That way the items won’t be removed from your library, just marked as missing/trashed. As soon as the media is available again it will recover (after a scan).
Obviously this isn’t a resolution for the issue being discussed, but it will mitigate some of the damage done by the files going “missing.”
3 Likes
