4.5+ GB written to cache per hour

[Unraidmule]

I am running Unraid v6.12.9

 

I only use the cache drive for docker.img and storing app data, so I can't figure out why it's writing data constantly 24/7. The actual total data used on the drive is static, and the drive is not actually filling up, just wearing the SSD endurance down. My guess is overwriting existing data with the same data?

 

The docker containers should only be writing log files, no transcoding or anything intensive takes place on the cache drive.

 

6.91 TBW @ 3,752 Power on hours

 

 

7.55 TBW @ 3,893 Power on hours

 

 

In 141 hours, 0.64 TB of data was written to the drive, which is roughly 4.5 GB per hour

 

 

 

[itimpi]

Have you disabled health checks for your docker containers as I believe they can cause excessive logging.

[Unraidmule]

3 hours ago, itimpi said:

Have you disabled health checks for your docker containers as I believe they can cause excessive logging.

Thanks for the reply. I have not disabled health checks. Is it a flag I have to put on each docker container?

 

I forgot to mention in the original post that the disk is also formatted btrfs if that helps diagnose the issue. 

[Unraidmule]

I added the --no-healthcheck flag to the containers which showed (healthy) in the advanced menu and now the writes have gone up to 5 GB per hour

 

Should I add this to all containers, regardless if they showed that state or not?

Edited April 3 by Unraidmule

[seestray]

8 hours ago, Unraidmule said:

I added the --no-healthcheck flag to the containers which showed (healthy) in the advanced menu and now the writes have gone up to 5 GB per hour

 

Should I add this to all containers, regardless if they showed that state or not?

it needs to be included in part of the docker image, if it doesn't show it in the adv menu, then that docker image doesn't have it.

 

I had a similar problem, it's been fixed for a while now - although can't remember what did it (there are some threads on here with potential solutions, btrfs or changing from the docker image to the folder structure might have helped - I recall changing those around that time). What file system are you running on the cache drive?

[Unraidmule]

3 hours ago, seestray said:

it needs to be included in part of the docker image, if it doesn't show it in the adv menu, then that docker image doesn't have it.

 

I had a similar problem, it's been fixed for a while now - although can't remember what did it (there are some threads on here with potential solutions, btrfs or changing from the docker image to the folder structure might have helped - I recall changing those around that time). What file system are you running on the cache drive?

 

Thanks for the reply, the drive is formatted to btrfs. I am also using extra flags on other docker containers to try to solve the issue, but it does not seem to have worked.

 

I added the --no-healthcheck to all docker images which were showing in the advanced menu — GluetunVPN must have a check because when I added it to that container it wouldn't start up.

 

For others, I added these flags:

 

--log-driver none --no-healthcheck --mount type=tmpfs,destination=/tmp,tmpfs-mode=1777,tmpfs-size=256000000

 

Still, the problem is occurring.

 

Is there a disadvantage of changing the docker image to folder structure?

[Unraidmule]

Does anyone else have any suggestions to fix the issue?

[Kilrah]

Try this:

https://forums.unraid.net/topic/136087-ram-disk-for-docker-statuslog-files/

[Unraidmule]

3 hours ago, Kilrah said:

Try this:

https://forums.unraid.net/topic/136087-ram-disk-for-docker-statuslog-files/

Thank you so much for the reply. This looks promising! Does it work on Unraid OS v6.12.9? Or the latest?

 

Currently, I am on v6.12.9

[Kilrah]

Yes, works on current

[Unraidmule]

If anyone reads this, using the solution Kilrah mentioned here:

 

 

I got the writes down to around ~2 GB per hour