Forge storage delete api not working?

is delete of forge storage api not working ?

    onClick={async () => {
      await invoke('deletekey'); 
                }}>delete all</button>
    onClick={async () => {
      await invoke('getallkeys')
                }}>get all key</button>   


resolver.define("deletekey", async () => {

  await storage.delete('testkey');

resolver.define('getallkeys',  async () => {

  await storage.query().limit(20)
      .getMany().then((result) => {

After perssing on delete button i press on get all keys button and i still get the ‘testkey’ with its value !
pleae i need some help


We have the same issue in our production applications. They were working fine for a year or more until this morning. In addition to delete operation, nextCursor value returned from getMany() is no longer undefined when there is no more results, it is now equals to 2 spaces, " ".

We are seeing this impact us as well.


From your code example, it’s not clear how your code expects to deal with eventual consistency. In other words, I think the fact that you still see testKey could be expected behavior.

Can you confirm that you never see the delete processed?

Definitely a “never” case and the key we provide was always correct.

1 Like

@denizoguz and @GirishReddy,

If you are seeing an incident, please report it to developer support. The ability of the Forge team to pick this up from the community as an urgent problem is also “eventually consistent”. Jokes aside, please do open a case with developer support.

I did that already. Thanks for your comments as well!

1 Like


Glad that you’re ahead of me! :wink: Did you make the report public so others can watch it? If so maybe let us know the issue key so the community can learn as case unfolds.

1 Like

I don’t think I can make it public yet, wouldn’t know where. I can only invite other users by e-mail. It’s ECOHELP-34613. I raised it with the type “Bug” instead of “Incident”. Is that a problem?

1 Like

As of now (2024-02-05, 7:21 PM CET) storage.delete is working again …

1 Like
1 Like

Since that incident is now marked as resolved, I’m going to lock this thread and repeat my advice about finding production incidents: please report them to developer support.

1 Like