Face recognition crashing the entire machine.
🗄️Server❓Question
Hi,
I recently started using immich for myself and it has been very nice.
For some reason though, no faces were recognised, so i did some digging and decided to press te renew(?) button for facial recognition in the admin panel and it put all my pictures into qeue and processed a couple, but then the dahboard froze and the keyboard attached to the server turned of and the monitor for it also froze.
The fans on the machine were also spinning at extra high rpm and cpu usage was around 98% on all cores when the screen froze.
I left the machine on when i went to bed since i figured it might just be a temporary freeze ,but around 10 hours after starting the task when i checked on it again the fans were still whinig and everything was still frozen, so i turned the machine off(with the motherboard powerswitch).
I then lowered the maximum ML task and tried again, but had the same error.
Specs:
16gb ram
I7 5820k(6 cores, 12 threads)
Gt710 to have display output, not accesible by immich.
I do not yet have a gpu installed to passthrough to the container, but i have a gtx1060 ready for when i have time to install it.
Are there any things i should configure otherwise?
Thanks in advance
I recently started using immich for myself and it has been very nice.
For some reason though, no faces were recognised, so i did some digging and decided to press te renew(?) button for facial recognition in the admin panel and it put all my pictures into qeue and processed a couple, but then the dahboard froze and the keyboard attached to the server turned of and the monitor for it also froze.
The fans on the machine were also spinning at extra high rpm and cpu usage was around 98% on all cores when the screen froze.
I left the machine on when i went to bed since i figured it might just be a temporary freeze ,but around 10 hours after starting the task when i checked on it again the fans were still whinig and everything was still frozen, so i turned the machine off(with the motherboard powerswitch).
I then lowered the maximum ML task and tried again, but had the same error.
Specs:
16gb ram
I7 5820k(6 cores, 12 threads)
Gt710 to have display output, not accesible by immich.
I do not yet have a gpu installed to passthrough to the container, but i have a gtx1060 ready for when i have time to install it.
Are there any things i should configure otherwise?
Thanks in advance
