Hey Team,
Every few weeks when I return to my Postman Flows, HTTP Request blocks show “Failed to load request” — even though the requests still exist in my collection. I have to manually re-select the correct request in every affected block before I can run the flow again.
Key details on the reference issue:
- Reproducible on desktop app
- Happens after a few weeks of not opening the flow
- Collection requests are intact; only the block-level references break
- Affects multiple blocks per flow, not isolated to one
- Every time need to select environment, variables in the request
Separately, I’ve also noticed that for deployed Flows on cloud, flows calling a module fail — even though the module itself runs fine standalone. The same module works in isolation but breaks when invoked as a subflow from a parent flow.
Together, these two issues make Flows unusable as a serious automation tool. For a small number of flows I can manually fix references and debug module failures, but scaling to 50–100+ HTTP Request blocks across multiple flows with module dependencies is not feasible when things silently break.
Could the team please investigate both issues and confirm whether fixes are planned?
SS of Issue 1 -
SS of Issue 2 -
Thanks for reporting these issues — sorry you’re encountering this.
Two questions; are you on the latest app version? Given that it happens after a while, I’m wondering if it’s related to app upgrade or if you’re on the same version between opens. Just fishing for a clue to help us diagnose.
On the second issue, this is usually an error in logic. Does the module for example fail to handle an error case, there’s some path that is usually not handled under composition. Have you looked at the run logs to drill into the module steps and see where the issue occurred? If it helps happy to look into this one together.
Thanks for the clarifications - we’re investigating the first issue.
For the failed module run, have you looked the run logs to drill into the module and see why it failed?
The run logs traces are also available in the flows analytics view if they were triggered automatically.
These should show you where in the module the error occurred and why.
One clue I see from your screen shot, your variables are not pinned — is the module invocation part of a loop? If so, recommend pinning the variables. It could be the module is not getting any input for example. If you can share more of the flow we can help you more effectively.
If the screenshot is more or less the entire flow, then please look through the runlogs and share what you find. If it’s faster feel free to dm or open a support ticket and we can take a look together.
HI @lunar-module-1 ,
Issue is happening whenever I am pushing new changes from my local to postman cloud. Because of this I have to reset everything again which is very frustrating.
- I create/update a flow merge it in my bitbucket repo.
- Push the whole workspace to postman cloud
- Switch to cloud
- Pull the whole workspace again
- Try to run Flow
- How my scripts work → Request Block → Common Module (HTTP request, returns pass/fail output using OR) → Evaluate block runs validation → Response/Display.
- Issue that I get is “An error occurred while executing the flow”
Main Flow/Test Script -
Common Module -
Hey @lunar-module-1 ,
Looks like the issue is persistent since the last time I spoke to you, Every time the app upgrades itself the reference to http request and variable are lost. Every time I have to connect blocks, update variables being passed which is really frustrating.
There is another issue that I observed after my colleague pushed his flow in the master branch and I took the pull. I couldn’t run his Flow because postman can’t find that module which was designed in my colleagues local branch. After doing some RCA we found out the YAML file has flowId mapped to the path of his local repo. I am really not able to understand why these issues aren’t solved in latest release. These issues have left me with no option but to look at some other tool/framework for my automation testing.
Your help here would be really appreciated!
I’m sorry you’re encountering this issue again. The paths should be relative to the postman project not absolute paths! Will work with the team to resolve this promptly
Hi @lunar-module-1 ,
Thanks for prompt response. Could you also confirm if the first issue I mentioned - Regarding lost references, connections, variables and env after each update or pulling/pushing to postman cloud can also be resolved. Because these are two different issues, resolution of one won’t make too much of a difference to my automation if I have to add connections between modules, update variables each time I or someone else do something.
The issues may be related. If you pull and push all your collections, environments and flows together, you should not be losing connections, and requests would not fail to resolve. If there’s a push/pull and resources are missing that can lead to the issues you’ve described. Since you’re not doing that, we have to look at the clues you provided and create a test case to see if it reproduces the problem. Will keep you posted. Of course we want to fix all the issues underlying the symptoms you described. The absolute path is def an issue.
Hey @lunar-module-1 ,
This morning app got updated again and I had the same issues. Let me explain what I noticed about diff env/setups -
- Postman Cloud -
- I did push the to postman cloud last week Wednesday.
- The app got a new update this morning. Postman v12.11.3 is the latest version.
Last checked on May 21, 2026 at 9:11 AM.
- I switched form local to cloud mode
- Scripts inside the flows section are failing.
- Observation - Even though env is pinned, connections are present between start and request (Of my common module) that I am calling fails.
- What I had to do to run it - Reselect the env again in the common module, common module returns the response.
- When I run the whole script again assuming that I should get the response from my common module it fails.
- Workaround - I have to create a new snapshot, select in the script and then it worked. Otherwise I get module not found error.
- Another observation - Deployed flow request worked without any issue without changing anything.
- Local -
- On local the issue is different. When I run the script it fails due to ““Error: An error occurred while executing the flow””, when I looked at the common module I am calling it had -
- missing connections, request failed to load but env was pinned.
- When I joined the connection between start block and request it worked.
- When I ran the script which is calling common module it worked.
Hope this clarifies the issue I am facing and your team will be able to resolve it asap.
Thanks for taking the time to share these detailed steps. We will def fold these into the testing and I’ll update our internal tickets.
We will have an update on this next week.
Hi @shubhankar-bhardwaj,
Thank you for the detailed steps. Our team has identified the issues and we are actively working on a fix.
We really appreciate your patience and for working with us to troubleshoot this. We will have more updates for you next week.
Hi @shubhankar-bhardwaj! Regarding the HTTP Request blocks losing their request references and variables, we haven’t been able to reproduce this on our end. Are you still running into this issue? If so, could you please share a screen recording so we can investigate further? Thanks!
Hi @shubhankar-bhardwaj ! Apologies for the silence on our end. Happy to let you know that both issues you flagged in this thread have been resolved:
- Module references breaking after a Git pull: fixed in v12.13.4.
- HTTP Request blocks losing their request references and variables: fixed in v12.14.5
Thanks for your flagging these issues and with the debugging you did, it helped pinpoint the root cause!
