Describe the bug
I had set up swarm between two bitaxe supra's a few weeks back, but ran into an unrelated issue that caused the IPs for the miners to change. Both devices would show their respective old IPs in the swarm page of the AxeOS web UI, but if I attempted to delete either IP from either device, it would fail. Looking at the request in the browser's dev tools, it was trying to send a modified json list of IPs to the "down" IPs, and never to the current one I was accessing. Manually sending the PATCH request using curl only to the new IPs, in my case with an empty json list, cleared the old devices.
To Reproduce
Steps to reproduce the behavior:
Have 2+ bitaxe devices in a working swarm
Do something that causes any/all of the device IPs to change
Attempt to delete old IP from swarm page
Expected behavior
I would expect that the API request would be attempted to all devices that the miner being managed is aware of, however I never see a request to the "current" IP of the miner. Perhaps a fail-safe when this condition is met would be to warn the user, and give an option to send the request to the current IP, but the other device(s) would need the same process run as well.
Hardware (please complete the following information):
Bitaxe HW version: Supra 401
Bitaxe HW vendor: Bitcoin Merch
ESP-Miner FW version: 2.1.3, 2.1.8
Hash Frequency: default
Voltage: default
Pool URL, Port, User: n/a
Additional context
Example curl request to send an empty list to one of the devices:
curl 'http://10.9.8.7/api/swarm' -X 'PATCH' -H 'Content-Type: application/json' --data-raw '[]'
WantClue
commented
1 month ago
Describe the bug I had set up swarm between two bitaxe supra's a few weeks back, but ran into an unrelated issue that caused the IPs for the miners to change. Both devices would show their respective old IPs in the swarm page of the AxeOS web UI, but if I attempted to delete either IP from either device, it would fail. Looking at the request in the browser's dev tools, it was trying to send a modified json list of IPs to the "down" IPs, and never to the current one I was accessing. Manually sending the PATCH request using curl only to the new IPs, in my case with an empty json list, cleared the old devices.
To Reproduce Steps to reproduce the behavior:
Expected behavior I would expect that the API request would be attempted to all devices that the miner being managed is aware of, however I never see a request to the "current" IP of the miner. Perhaps a fail-safe when this condition is met would be to warn the user, and give an option to send the request to the current IP, but the other device(s) would need the same process run as well.
Hardware (please complete the following information):
Additional context Example curl request to send an empty list to one of the devices:
curl 'http://10.9.8.7/api/swarm' -X 'PATCH' -H 'Content-Type: application/json' --data-raw '[]'