Environment deletion fails with FF CLI built from main

[peterbroadhurst]

This is a bit challenging in a migration scenario, as I would expect the normal flow is:

• Install new ff
• Find you need to delete your old dev env
• Run ff rm (currently fails)
• Re-run ff init

./ff/ff rm -f dev                    
deleting FireFly stack 'dev'... Error: Error: No such volume: dev_geth

Usage:
  ff remove <stack_name> [flags]

Aliases:
  remove, rm

Flags:
  -f, --force   Remove the stack without prompting for confirmation
  -h, --help    help for remove

Global Flags:
      --ansi string   control when to print ANSI control characters ("never"|"always"|"auto") (default "auto")
  -v, --verbose       verbose log output

[peterbroadhurst]

Hmmmn, well I also can't create a new environment I just created...

[peterbroadhurst]

➜  firefly-cli git:(main) ./ff/ff init --multiparty=false gw2 1
initializing new FireFly stack...
Stack 'gw2' created!
To start your new stack run:

ff start gw2

Your docker compose file for this stack can be found at: /Users/pbroadhurst/.firefly/stacks/gw2/docker-compose.yml

➜  firefly-cli git:(main) ./ff/ff rm gw2                       
WARNING: This will completely remove your stack and all of its data. Are you sure this is what you want to do?
completely delete FireFly stack 'gw2' [y/N] y
deleting FireFly stack 'gw2'... Error: Error: No such volume: gw2_geth

Usage:
  ff remove <stack_name> [flags]

Aliases:
  remove, rm

Flags:
  -f, --force   Remove the stack without prompting for confirmation
  -h, --help    help for remove

Global Flags:
      --ansi string   control when to print ANSI control characters ("never"|"always"|"auto") (default "auto")
  -v, --verbose       verbose log output

Error: Error: No such volume: gw2_geth

[peterbroadhurst]

Ok - with a bit more testing, I think the scope of the issue is:

• v1.2 environments
• v1.3 environments that haven't been started (init run, but not start)

[nguyer]

Thanks for opening this! I think I have seen this issue as well. I thought I fixed it, but maybe it was a similar but slightly different issue.

For background we used to swallow a lot of errors, and when we added the linter it wasn't a big fan of that. Making the linter happy ended up with some errors that still should have been swallowed (or at least logged) getting returned all the way up. This was one of those. Should be an easy fix. I'll open a PR for it.

[nguyer]

@peterbroadhurst Are you sure you are running the latest code in main? I think this existing check should allow this condition to not return the error:

firefly-cli/internal/stacks/stack_manager.go

Lines 707 to 709 in 3679daf

	if !strings.Contains(err.Error(), "no such volume") {
	return err
	}