Immich server becomes unresponsive during uploads (can’t SSH in)
❓Question🗄️Server☑️Ready
Hi all,
I’m having stability issues with my self-hosted Immich setup. After some time (often when uploads/jobs are active), the box becomes unresponsive: Immich UI stops loading and I can’t SSH in either. Sometimes it comes back on its own later, sometimes I have to reboot/restart things.
It's been running just fine for almost a year now, but it's recently become unstable, it seems to become unstable when my mother uploads images from her library and jobs run.
My library is large-ish, 120k images, 12k videos ~1.5TB
It’s running inside an LXC on Proxmox on an N97 Mini-PC with 16 GB RAM total, and I’ve allocated 12 GB to the Immich container and 3 CPU's.
The PostgreSQL database is stored on the local SSD in the Proxmox host, while the photo upload library itself is stored on a separate TrueNAS system (HDD pool) mounted over NFS on a 2.5GBps wired connection.
I think the ML jobs are somehow killing the server, I've tried fixing it but no luck. Not sure why the whole server would become unresponsive, I can't even SSH into the LXC
I've attached my docker-compose and logs
I'm on v2.5.6
I’m having stability issues with my self-hosted Immich setup. After some time (often when uploads/jobs are active), the box becomes unresponsive: Immich UI stops loading and I can’t SSH in either. Sometimes it comes back on its own later, sometimes I have to reboot/restart things.
It's been running just fine for almost a year now, but it's recently become unstable, it seems to become unstable when my mother uploads images from her library and jobs run.
My library is large-ish, 120k images, 12k videos ~1.5TB
It’s running inside an LXC on Proxmox on an N97 Mini-PC with 16 GB RAM total, and I’ve allocated 12 GB to the Immich container and 3 CPU's.
The PostgreSQL database is stored on the local SSD in the Proxmox host, while the photo upload library itself is stored on a separate TrueNAS system (HDD pool) mounted over NFS on a 2.5GBps wired connection.
I think the ML jobs are somehow killing the server, I've tried fixing it but no luck. Not sure why the whole server would become unresponsive, I can't even SSH into the LXC
I've attached my docker-compose and logs
I'm on v2.5.6
