Stale data returned in Secrets automation flow

rtaraniganty

We are a business customer with a few users and we wired in 1Password's secrets automation flow into our test/deployment processes. We recently started noticing that we are receiving a "stale" version of the secure note from the vault we granted access to. Neither a new secure note (or any item) nor the changes to an existing note are being served.

I can see the new information/items in the UI, but the API doesn't return the expected values.

I sent an e-mail to 1Password support and immediately received a nice and friendly automated e-mail with a case number. Since then there's nothing from 1Password's side. I have replied to the e-mail and provided a couple of updates. There's no response or even an automated ack.

---

1Password Version: Not Provided
Extension Version: Not Provided
OS Version: Not Provided
Referrer: forum-search:Is 1Password's e-mail support completely broken?

David_ag

Hi @rtaraniganty -

I work on the integrations team that developed 1Password Connect.

Huge apologies for the delay in getting back to you. I saw my colleague has engaged with you in our support system. I'm going to take a look at the issue as well and continue the conversation over email.

Best,

David

joe1698

I am also having this issue. What was the resolution for this?

ngustafson

I am having the same issue as well.. Any resolution yet?

LuukM

Are you using the Docker Compose image as a Connect server?

What i found, was that there is not really a functional sync in place. It syncs everytime the container / volume is created, not when it is restarted.

What helped me, was removing the container and volume (docker volume rm {container}). This will force a new sync, giving you all the latest items

Joris_1P

Hi all!

My apologies for the late response here. We are doing a deep investigation into what is going on here. So far, we have not been able to reproduce the issue, but with some extra info we may be able to.

So could one or more of you help me with the following points:
1. In your case, is Connect running continuously or is it started just before it is needed?
2. Do you have any logs of the containers (both connect-api and connect-sync) for a time that you ran into this issue? If possible, from the moment Connect started until you made a request that returned stale data.
3. Have you always had this problem? Or did it only start showing up after a while? In case of the latter, do you have an indication of how long it took for the problem to manifest itself?
4. How are you running Connect? (docker-compose, AWS ECS, k8s/Helm, etc)
5. Which version of Connect are you running?
6. Which Connect integrations (SDK, Terraform provider, Ansible plugin, etc, or just plain cURL) do you use?

Answers to these questions could possibly help us reproduce the issue

In the meantime, we are also working on some more general improvements to the syncing mechanism, that should make it more robust.

Joris

Joris_1P

Hi folks!

Quick update from my side: earlier this week we were able to reproduce this problem and we have released v1.5.1 of Connect, which should fix the problem we found.

You can update by either running docker-compose pull or updating the image's tags to 1.5.1. We have also applied a server-side mitigation that should resolve the issue even for older versions of Connect, but you may have to make a change to an item in one of your vaults to trigger a new sync and get all content in Connect back up-to-date.

Let me know if that resolves the issue for you.

Joris

LuukM

Hi Joris,

Not to hijack the thread, but for me, this seams to solve the issue of stale data!
Cheers!

Joris_1P

Thank you for letting me know. I am glad to hear that!