set logpush job id to "" when not found #798
Conversation
|
Thanks for putting in the time to fix this one 🍰 As the scenario we're addressing here is when the logpush job isn't in the remote but still in the state, we should include a fix for #795 where attempting to delete the resource fails due to it not being present in the remote. This should then cover off all the cases where the remote is out of sync with the local state. |
|
Good catch @jacobbednarz, I must be blind because I completely missed that issue. For now this small patch solved our immediate issue and we're just running a custom build of this branch internally for the time being. I won't likely have the time in the next few days to address anything additional for #795 so anyone else is more than welcome to. |
Should someone delete a resource outside of Terraform, it won't be found when the remote delete is called. Instead of throwing an error stating we couldn't find it, just remove it from the state. Fixes cloudflare#795
| return fmt.Errorf("error deleting logpush job: %+v", job.ID) | ||
| } | ||
|
|
||
| return resourceCloudflareLogpushJobRead(d, meta) |
There was a problem hiding this comment.
We don't need to call Read as there isn't anything to "sync" after we d.SetId("").
fixes #797
There are currently no tests for the
logpush_jobresource that I could find so I haven't added or amended any. I assume that is because it requires resources external to Cloudflare?Either way I have build the provider with this change and tested it ourselves and it's working as intended.