I have been having an issue with deploying my changes in pipeline builder getting the error: “An unexpected error occurred. This may result from a system outage or misconfiguration”. It seems that its failing to begin deployment rather than an error causing it to fail. Has anyone seen this before? - I haven’t seen anything in documentation or community forums.
I have already troubleshooted with AIP Assist, attempted to deploy the last successful deployment (failed) and deploy smaller changes at a time. Do you have any recommendations on how to resolve it or where more extensive documentation to resolve it is?
Hey! Just to confirm, there are no pipeline errors correct? Is this is the first time you’re trying to deploy your pipeline and if so do you have the right permissions to build it? Is this happening with just one pipeline or any pipeline on your enrollment
I have deployed this pipeline multiple times over the last month or so with no issues but have now been getting this error since last Wednesday on the pipeline. The issue is specific to this pipeline and when asking others with access to it, they are having the same issues deploying.
I can confirm there are no pipeline errors. It appears to not even begin deployment and instead presents error described above (for ease, this error is: “An unexpected error occurred. This may result from a system outage or misconfiguration”).
Would you be able to get an error id or trace id? You can get this by right-clicking your screen and selecting Inspect. Then , find the failing request (with the red X) in the Network tab, and click into the error and copy the errorInstanceId from the Response tab
Hey there Helen,
I have been receiving the same misconfiguration/system outage error when deploying objects. this is the error code I get.
“type”: “STATUS”,
“status”: 500,
“body”: {
“errorCode”: “INTERNAL”,
“errorName”: “Default:Internal”,
“errorInstanceId”: “d628e4ec-08da-44a5-92be-7a943abfb73d”,
“parameters”: {}
Hey! Unfortunately I wasn’t able to find @amhall’s error, but for @Maverick, the root cause is that there are too many ontology objects already registered in your ontology. For dev tier, the maximum number of object types which you can create is 60.
We recognize that this behavior is confusing, so we’re tracking a way to make this error more readable from our side!
Hi! Is there something else I can do to help you identify this error? I am keen to get it working again.
I have tried deploying again. See below the full error response:
{
“errorCode”: “INTERNAL”,
“errorName”: “Default:Internal”,
“errorInstanceId”: “f2e5ddf1-63f9-4a1b-b8b7-6451e25a3714”,
“parameters”: {}
}
The console is actually describing error “Failed to load resource: the server responded with a status of 500 ()”. Not sure if this gives any further insight as all other Palantir apps are working fine and it seems to be an issue specific to this pipeline not a user.
Hey @amhall sorry for the delay here, we were having trouble finding your specific error instance. Just to double check, you have <60 objects in your ontology?
Hi @helenq, there are over 60 object types in our company ontology but the pipeline is only connected to a handful of these. I have removed these objects but its still coming up with an error albeit a different one. It is saying: “Some output datasets already exist. The output dataset names should be unique in the output folder.” I have checked files and cannot see any overlapping names or RIDs. Please recommend a next step.
In future, how would you recommend storing and building additional use cases out if we are limited to 60 objects?
Hey @amhall, do you have other resources with that name (not just dataset files?) Can you try renaming those datasets and seeing if that helps
How many use cases do you currently have and how many objects are you using for each one? Right now 60 is the limit for our dev tier accounts so curious to get more dots on what leads to the most object creations (# workflows, test objects, etc.)
Hi @helenq , thank you for your help - a tidy up of the ontology and datasets in files plus deploying on a new branch has seemed to get the deployment issue fixed. Thank you so much for your help!
Regarding the use cases etc - we are currently working on 2 use cases, one of which has quite a complex workflow. There are definitely some aspects that I’m sure we can streamline as we become more familiar with the apps and learn some best practices though.
