I’m facing an issue. I have two devices on a local network: data_server and client. I’ve set data_server in ‘send-only’ mode and client in ‘receive-only’ mode. Furthermore, my data_server has a vast number of files and directories. After syncing for a week, I noticed that my client has created all the directories, but no files have been synced as per the GUI. When I checked the sync directory on my client, I found many hidden files waiting to be renamed.
Could someone guide me on any settings or adjustments needed to resolve this? I’m using Syncthing version 1.23.6 For Linux,and my os is Centos7. If more detailed images or information are required, please let me know. Thank you!
hello gadget
Thank you for the reply, gadget. In my “Out of Sync Items”, there are numerous items waiting to be synced. However, on my server, I can see that they are hidden files. How can I investigate and resolve this issue?
The hidden files created by Syncthing are a normal part of the sync process. From the screenshots, they look fine, so no need to worry about them for the time being.
On “server2” you have the file pull order set to “Largest First”, so in the screenshot labeled “sync1”, all of the smaller files are queued up waiting for the larger files to be downloaded first. In the “Out of Sync Items” panel, jump to the end of the list and see if there’s an error syncing the largest file (if there is an error, hover the mouse pointer over an item to see the full message).
I have files that total around 7-8GB in size, and the synchronization process has been ongoing for over a week. Currently, I don’t see any upload or download traffic in Syncthing. I’m unsure if the issue is due to the large number of files I have. My data_server exhibits a lot of read-write operations on many files during synchronization. Could this cause any issues?
A large number of files – in your setup, over 10 million – will result in a lot of database activity for Syncthing. If Syncthing’s database is on the same storage, then it’s going to be an issue (especially for HDDs). CPU and RAM usage also become more of a factor. Disabling file watching on “data_server01” and relying only on full scans also adds more processing overhead. Others who have posted about having a million+ files have reported the initial sync taking days. However, it generally should just result in slow syncs, not completely stall.
Right now I see two immediate options:
Find the largest file in the list of out of sync items and see if there’s an error message for it. Either scroll thru the “Out of Sync Items” panel, or enable more verbose logging and review the log output.
At least temporarily change the pull order on “server02” from “Largest First” to “Smallest First” to see if Syncthing quickly transfers a bunch of small files. As long as it’s set to “Largest First”, you don’t know yet if that one large file is holding up everything else in the sync queue.
