Trash does not go away in 7 days

Node Operators troubleshooting
148

Seems the current workaround is:

149

I have the same problem on two of my nodes

Is this solution something permanent or something temporary?
Can I leave this alone or do I have to change lazzy walker to true after a while?

Would this be the correct configuration?

pieces.enable-lazy-filewalker: false
storage2.piece-scan-on-startup: true

150

something temporary

Yes and you may change it back, when it would be fixed. You may track the release changes either on our GitHub or here, Search results for 'release preparation order:latest' - Storj Community Forum (official)

yes

151

I have found a bug in my storj node.
The trash reported from the dashboard is 194.72GB in size, but on the filesystem the trash folder is: Size 43.39 GB (46,589,713,717 Bytes). 252050 Files, 4163 Folder containing.

The blob folder is 0.8TB in the dashboard, but in the filesystem it is Size: 749.56 GB (804,830,621,172 Bytes). 3866749 Files, 4100 Folder
I think, this is ok, this referes to the rounding of the dashboard.

So, what can I do to correct the trash size in the dashboard?

The node is running at version 1.102.3 since 360h. no other bugs detected. (Default Docker container)

Instead of reastarting the docker with the lazzy filewake. Is there a possibility of running a shell command in the docker directly without restarting and loosing downtime?

152

Probably this is what you need to do:

1 Like
153

Yep, the workaround posted by @jammerdan should help.

154

But there is no shell command to do that inside the docker container?

Which would be the docker ENV variables for that configuration, or will that problem be solved in 1.1.04.3, and if yes, when will the update be out?

155

You can. Just put these options after the image name, i.e.:

docker run .... \
storjlabs/storagenode:latest \
--pieces.enable-lazy-filewalker=false

I hope so. I cannot track, because there is no issue on the GitHub.

156
2 Likes