Is there any way to identify what is running in background every ~5s and uses/writes my ZFS UserData and network?
I have NextCloud 30.0.3 installed over TrueNas Scale ElectricEel-24.10.0.2 (as official App plugin)
My Apps inside TrueNAS are set already to “BootPool” (SSD) and NextCloud AppData Storage & Postgres Data Storage is inside iXvolume.
Only my User Data Storage is redirected to my nas ZFS raid ( which uses physical hdd disks ) …and here is a “problem”
I can here every ~5 seconds that my disks write something and also there is network traffic visible inside TreNAS info next to app info ( which I think is related to this activity )
But why does this movement cause writing to the disks if they are not in use (no external application/client uses writing) - I am the only user of this instance. Any help where to look?
Is there outgoing network traffic as well? If it is a client, you should see it in the webserver logs as well. And then go back to the client if there is one.
For Nextcloud server itself, there is a cronjob running regularly, but it should be 5 min by default (was 15 min before). You can adjust these times to your use case (e.g. less often at night when there is little activity).
Yes, I have “in” and “out” traffic ~ [11,1kb/s / 11,kb/s]
This is why it is so strange…
Webserver should use UserData “data” ? ( I thought WebServer only reads/writes to files in AppData and uses the above. Why would it use UserData? )
Also I’m not using crone ( default AJAX setup here ) - as I’am only user.
I had already thought about this before. But unfortunately. It turned out that the default is AJAX.
A health check which is carried out at regular intervals. This is common when using Docker images so that you can see whether the container is running successfully.
It sounds reasonable, but for several applications only the NC instance performs this action, no other. (and I have two media servers next to NC) So it’s a matter of NC directly (and its Docker settings) - and I still wonder why it “queries” the UserData space and “moves” physical disks
I’m seeing something like this from my Nextcloud install in Truenas Scale.
This error/warning hit my Nextcloud log every ten seconds and doesn’t stop. I see this in a real time scroll. This is a screen capture from the 5th that I used asking in the Truenas forums.
Yep! This is is It as I assume …
Had to move all App folders to second pool.
I’ve added additional storage to TrueNAS as vDEV drive from my Proxmox.
After this I moved all my Apps to this pool. So not using any HDD for App configs any more, just for Data. ( there is no more noise )
Thanks for the reply! I’m going to have to look into doing that. I just have the single pool at the moment. It’s just not about the noise but mostly the IO hitting all 5 of my drives (raidz2) at the same time every 10 seconds.
Any pointers about moving the Nextcloud app. ( the only app I use)
I have made a fresh install for all my apps on my new pool ( inc. Nextcloud )
I found quite nice video for latest version of NextCloud - and commented as well because my NextCloud app v:30.0.3 looked a little different and I get an error at the end of install …
I had to delete Jellyfin instance from my TrueNas - It’s possible there was some conflict in the Docker instances. And had to purge all older image containers (No big deal - I’ll get the rest back no problem) Thanks to this, NextCloud installation was completed successfully (it didn’t want to install before - unfortunately I didn’t save the errors) And as a curiosity, I did not set any special permissions for Dataset - the installer replaced them itself. So it works now. I just can’t figure out what process is writing some data to the disks every few seconds ~4s. (I’m using zRAID5 for UserData, and the rest is on BootPool, including “ix” containers). Something is clearly communicating because I can even see network traffic.
Thanks for the reply! I may give this a go. I was hoping moving/reinstalling Nextcloud would “cure” the endless writing to disk issue, but maybe not? Unless this behavior is normal?
From what I searched for and from the knowledge I managed to gain, “this is how it has to be and it is quite normal”. …and I searched for a long time
As I described above, only this solution brought the desired result for me.