Once your node updates to v1.12.6, this should no longer be an error that occurs. We still have improvements to make, but at least you won’t be blocked from sending orders.
3 Likes
is 1.12.6 a future release or a replacement for 1.12.3
1 Like
It may be not the same issue but may be related to the issue here. I started a node on a router ago. Running node on OpenWRT router?
After I updated it to 1.11.1 I got activities decreasing on all (ingress, egress, repair and audits) until it was 0 on the day after.
I didn’t know the architecture on how it work and still don’t know it very well but my main OS drive on my router the space remaining at that time was 0 even if I put the storagenode on the external drive. Reading here that orders folder is not where my storagenode folder data are make the network not working has supposed. The orders use spaces on my main drives OS and when full things may go wrong. My node are still online, but nothing else append. I said it may be related it’s because it append at the same time with the issue here. I only have some MB of space on my OS drive. It is not expected with the most storage node operators here.
EDIT: Updated to 1.12.3 now
The orders folder usually created in the folder with config.yaml. If you used a binary from a docker image, then check where is orders folder is located.
You can stop the storagenode and move orders to the disk with remained storage and specify the path to it in the config.yaml, for example:
# path to store order limit files in
storage2.orders.path: /root/.local/share/storj/local-network/storagenode/0/orders
3 Likes
Do you see such errors after the last update?
The 18th of september the node had a power suply problem, since that day it has many errors in the log.
each 5 minutes "ERROR orders listing orders {“error”: “order: unexpected EOF”
also, each 5 minutes “ERROR contact:service ping satellite failed”
orders/unsent folder has 1050 files
I tested databases and all are ok
Get-ChildItem f:\nodo1*.db -File | %{_.Name + " " + (sqlite3.exe $_.FullName “PRAGMA integrity_check;”)}
The node is in windows 10 and it has about 3TB of shared data in the node
Thanks a lot. 
Now i move the 2 corrupted unset files to other folder(in previus comment said they expire.
I have 10 files in unsent folder from today but there aren’t new erros in the log file.
(it would be very usefull to have the id file in the log line error)
I hope the transfers file will increase again
I have 2291 unsent orders. This could be a long night. Can I suggest a reference in the error code to identify the corrupt order file?
What is the version of your node ? What does your log show for
I have watchtower so same version as everyone else. 1.13.3. Same error. Orders EOF error. The node seems to be running ok but obviously is not sending orders.
Have you tried removing the oldest order files ?
Well it’s been an interesting hour. That latest update must have fixed the issues because I’ve been sitting here watching the unsent orders slowly reduce even though I was still getting the EOF orders error. Now I only have new unsent orders and for the last 15 minutes no EOF orders error.
3 Likes
You are correct, one of the v1.14 point releases had a quick patch which handles the unexpected EOF situation. A further improvement (a new file format for the orders file) is expected in v1.15. See this comment: Error: "order: proto: pb.OrderLimit: illegal tag 0 (wire type 0)