-
Notifications
You must be signed in to change notification settings - Fork 79
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Damselfly keeps crashing #528
Comments
Are you able to attach the image it's processing when it fails? Or can you email it to me at [email protected]? |
I've pushed a potential fix for this (entirely speculative as I can't repro). Can you give it a go by pulling the dev version (pull |
@Webreaper I'm not sure which image its failing at, To make sure I send you the right one, in the logs I posted would it be the last image it couldn't find thumbnails for?
This is another error I've seen often, I believe this happens when ever I refresh the web page for damesfly when its going through my reverse proxy.
I will try the dev version and let you know, there are still around 20k images left to process for thumbnails. |
With the dev version I'm not seeing a crash after about 15 minutes of it running. I am still seeing the errors about the thumbnails but the application doesn't seem to be getting killed. I'm also seeing thumbnails loading on the web page. A bit more detail, I'm using a reverse proxy (nginx) to access damesfly through my public domain. It seems like something about that setup isn't working absolutely perfectly. In addition, the path for all my photos to damselfly is a NFS share. The images aren't physically on the same hardware as damselfly. The images are on another server sitting right next to the server damselfly is running on. There's 4x 1Gbs network connections between the two systems. Here's the log I'm seeing now.
The HTTP and Weird thing now is even though thumbnails are loading, if I try to open the image by double clicking I get an "image not found" error. Edit: Updated too soon, it did end up crashing again.
Another update, sorry for all these edits. This time it looks like some resource usage caused the whole VM to lock up, when I started damsefly up again I saw a new error.
|
Not sure how much I can help with the proxy thing. Never had that reported, and I won't be able to repro. For the crash, can you send me _DSC1897-HDR.dng, _DSC2181.dng, _DSC6523.dng and _DSC6523.dng so I can test why they're causing the crash? (or just send one, and then I can see if i can repro). |
No worries on the reverse proxy thing, I can troubleshoot that on my own. I've emailed you the requested pictures, its only 3 so I don't mind sending you more data to test with. |
After a few hours it seems like a memory leak crashed most of the VM. The last logs in netdata show a spike in memory usage, and proxmox shows a ton of disk writes before the crash. The last logs are
Tried this again with 24GB of ram to the VM and same thing happened. |
@Webreaper any luck in recreating this? |
Sorry, stuff cropped up so haven't had time. Still on my list to investigate. |
No worries, let me know if you need anything else when you get to it. Thank you. |
@Webreaper Throwing more RAM at it seems to have solved the issue. I gave the VM 24GB and added another 64GB to the host so there's more room for all the VMs running there. Damselfly seems to have finished all the indexing tasks for the 7000 new images I've added since the issue started. I'm letting it do the AI processing now. |
It's a bit hungry. 😁 |
Damselfly version: v4.1.3.0
docker compose
I added my whole 1.8TB photography collection to damselfly and I keep seeing messages like what you see below. I have AI object detection and face recognition turned off. Thumbnail generation is running.
Damselfly is running in docker, inside a Debian based VM in proxmox. The VM has 20 cores and 15GB of ram assigned to it.
Do these errors just mean damselfly doesn't have the thumbnails for the images yet? Is that related to the 'killed' message, and what seems like a crash. Ram usage is very low, only about 9% of total ram is being actively used.
The text was updated successfully, but these errors were encountered: