diff --git a/_docs/developer/getting_started/worker_vm.md b/_docs/developer/getting_started/worker_vm.md index b939ab99..da3a91d0 100644 --- a/_docs/developer/getting_started/worker_vm.md +++ b/_docs/developer/getting_started/worker_vm.md @@ -27,20 +27,28 @@ machines* in addition to your primary vagrant virtual machine. ``` For multiple workers, use the `-n` flag. (ex. `-n 3` for 3 machines). + This command will generate workers for `virtualbox`, or `qemu` if you are on M-series macOS. If using a provider other than the default for your system, use the `--provider` flag. - (ex. `--provider parallels`) + (ex. `--provider my-custom-provider`) -4. If you are on MacOS running QEMU, make sure to restart the network socket. +5. If you are on MacOS running QEMU, make sure to restart the network socket. ``` vagrant workers socket restart ``` If the VM runs into errors when attempting to reach the internet, try starting the socket - in public mode. (`--public`) + in public mode. (`vagrant workers socket restart --public`) -5. Now you can provision the virtual machines with: +6. Now you can create the worker machine(s) with: ``` vagrant workers up ``` + Do not use the --provider flag with this command, since it will conflict with the + provider generated in step 3. + +7. Once all the workers are fully set up and running, `vagrant ssh` into the main VM and run: + ``` + refresh_vagrant_workers + ``` --- @@ -50,14 +58,15 @@ If you would like to ensure the worker is functioning properly, or enter the wor To connect to a worker machine through SSH, run: ``` -vagrant ssh +vagrant workers ssh ``` If you want to test the connection between the primary VM and a worker, you can first `vagrant ssh` into the primary machine and then run this command to SSH into the worker from there: ``` -su submitty_daemon -c ssh submitty@ +su submitty_daemon -c ssh ``` -The IP address of the worker machine will be indicated in the `.vagrant/workers.json` file. + +The list of worker names can be displayed with `vagrant workers status`. __NOTE__: Depending on the performance of your computer and the size of the autograding queue passed to the worker, the SSH command may hang for some time.