June 1, 20206 yr I have googled and "ducked" this issue but could not find relevant information on the subject... My array, after 27 day of uptime had around 900 000 000 (nine hundred million) writes on the cache and about 40 000 on the long term storage... If this data is real, this is going to kill my cache drives really quick! I have just restarted the array for some reason and stopped every VMs and dockers to check if writes will still be high (nextcould, mariadb, PiHole dockers and an Ubuntu VM for my HTPC setup). This screen capture is after about 5 minutes of doing nothing on the array! Is this normal behavior? Am I right to be worried? - Normand
June 1, 20206 yr Community Expert Seems to have something to do with btrfs and/or the official plex docker container: https://forums.unraid.net/bug-reports/stable-releases/683-docker-image-huge-amount-of-unnecessary-writes-on-cache-r733/?tab=comments#comments
June 1, 20206 yr Author 4 minutes ago, strike said: Seems to have something to do with btrfs and/or the official plex docker container: https://forums.unraid.net/bug-reports/stable-releases/683-docker-image-huge-amount-of-unnecessary-writes-on-cache-r733/?tab=comments#comments Thanks for your fast response... I don't use PLEX, but I chose btrfs for the system... I am in the process of rebuilding 3 computers at once tonight so I can't read the whole thread... did You find a solution to this issue that you can point me to?
June 1, 20206 yr Community Expert I haven't read the whole thread myself only skimmed through it. Seems to be a couple solutions, one of them is to reformat the disk to xfs. But then of course you loose the ability to have a cache pool. The other possible solution was something about modifying a file related to docker and creating a dedicated share... You have to check it yourself. I've had this issue myself for a long time. I managed to reduce the writes by modifying my ombi docker container though, but I still have 170 Million writes in 70 days. So I'm planning on re-formatting the disk to xfs since I only have 1 cache disk anyway. Maybe Limetech comes up with a fix (I hope) Edited June 1, 20206 yr by strike
Archived
This topic is now archived and is closed to further replies.