After updating to V6.x I see increase of the power consuption, meassured with a shelly. After rollback to 5.18.x power consumption fine again.
see picture below. approx 9:00am I have upgraded to V6.x. After 18:00 I rolled back, and all fine again.
does anyone know how to prevent this? Or anyone have the same?
What are the units on the vertical axis?
Have you investigated which processes on the ProxMox container are using resources that would result in power draw increase (CPU, etc.)?
The verical axis is in Watts. you see above, before with FLT V5.x the power consumption is about 8-10W. After installing FLT V6.x it increases, peaks up to 20W frequently. after rolling back, you see again, back to normal. So it is clearly the piHole container. I have checked with htop, there was a process going up and down flt something. not sure why it increase that much. Even i had no open browser window to pihole server.
How big is your Pi-hole query log, located at /etc/pihole/pihole-FTL.db
oh, 4.6 GB
how can we prevent getting that big?
That's likely the source of your problem. In V5, the default number of days stored in the query log was 365 days. In V6, it is 91 (it won't trim your older entries from V5). But, in either version, this is a user controllable option.
Run these commands to move the existing query database to a new file name and create a fresh database.
sudo service pihole-FTL stop
sudo mv /etc/pihole/pihole-FTL.db /etc/pihole/pihole-FTL-old.db
sudo service pihole-FTL start
alright, thank you. Will try and report back if any further issue should occur.
looks better now,Thanks for your help. But still a little more power consumption (+~ 0.5-0.7W), I can live with it. But how can we reduce it back as it was before. Why is it sucking now more power as before? See screenshot
htop:
pihole-FTL is "flipping" to 2.3% every 3-5 seconds (might be ok).
That CPU has a TDP of 28W. 0.5W is nothing.
How frequently is htop updating? Are you taking in to consideration that FTL now is the web server where version 5 required ligttpd?
That consumption screenshot doesn't really show me anything because it's a wide band of noise. The max consumption doesn't look any higher (the top of the band and peaks). If you want a true measure of the differences then you need to check the kWH and not something like just W.
