I suppose that many people have quite some unused space on network folder. It would be quite useful to be able to store the nodes on an nfs or samba folder. Is there any plan to do so?
If you are using antctl on Linux or Mac for example you can tell antctl to use a different directory with the --data-dir-path option.
Or just mount the share to the appropriate place in the filesystem at the location it expects to use. eg.
/home/safe/.local/share/autonomi
and I think that would work for Launchpad as well on Linux and Mac. But maybe with a different directory. I’ve not used Launchpad for a while so I don’t know if it’s a different location. Someone else will.
I think you would struggle if using Launchpad for Windows because network shares appear as drive letters in Windows and Windows Launchpad will try to use storage under C: in the user’s directory.
I doubt they will add the ability to choose a mounted drive because Launchpad is meant to be a really simple tool that just runs and adding the ability to choose other storage locations would break that simplicity and add in the complexity for users of finding out whether any availability or performance issue is due to the NAS, network, permissions, etc.
It’s much simpler for the simple tool to only allow the ability to use storage that is known to be attached and performing well because it’s definitely attached, writable by the user and performing well enough because the user is already using it and the software was launched from it!
It’s reasonable to assume that anyone who wants to do something a bit out of the ordinary like using network storage is capable of using the tool with more options - antctl.
This was so useful. I think it works.
Actually it doesn’t. The new nodes won’t start:
Attempting to start antnode32…
Failed to start 1 service(s):
✕ antnode32: The PID of the process was not found after starting it.
Error:
0: Failed to start one or more servicesLocation:
ant-node-manager/src/cmd/node.rs:875Backtrace omitted. Run with RUST_BACKTRACE=1 environment variable to display it.
Run with RUST_BACKTRACE=full to include source snippets.
I also have 3 faulty (botched upgrade process) nodes which I cannot remove with the same error.
I think what happens is that there are still services registered and trying to run which are hogging those service names. I think if you do a antctl reset and then reboot to clear out the services things will work again. You can also manually remove the services but it will be quicker and easier to reboot. Especially if you are on a Mac.