Skip to content

Commit dd058a4

Browse files
Paul-CornellPaul-Cornell
authored
Additional troubleshooters (#788)
1 parent f32eeed commit dd058a4

File tree

6 files changed

+188
-2
lines changed

6 files changed

+188
-2
lines changed

‎docs.json‎

Lines changed: 7 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -536,14 +536,19 @@
536536
"group": "Issues",
537537
"pages": [
538538
"support/issues/overview",
539-
"support/issues/api-connector-secrets",
540539
"support/issues/authorization-permissions",
541540
"support/issues/configuration-resource",
542541
"support/issues/quota-billing-rate-limiting",
543542
"support/issues/network-connection-timeout",
544543
"support/issues/data-format-schema-validation",
545544
"support/issues/document-processing",
546-
"support/issues/internal-file-handling"
545+
"support/issues/internal-file-handling",
546+
"support/issues/api-connector-secrets",
547+
"support/issues/cannot-locate-credentials",
548+
"support/issues/get-authenticated-user-error",
549+
"support/issues/workflow-job-in-progress",
550+
"support/issues/no-fast-partitioning-for-images",
551+
"support/issues/google-drive-schema-validation"
547552
]
548553
}
549554
]

‎support/issues/cannot-locate-credentials.mdx‎

Lines changed: 36 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,36 @@
1+
---
2+
title: S3 destination connector - Unable to locate credentials
3+
---
4+
5+
## Issue
6+
7+
When you try to run a workflow in Unstructured, and the workflow contains an Amazon S3 destination connector,
8+
you get an error that contains text similar to one of the following, and the workflow fails to run:
9+
10+
```text
11+
500: Error in uploader - [NoCredentialsError] Unable to locate credentials
12+
```
13+
14+
```text
15+
failed to invoke plugin: Unable to locate credentials
16+
```
17+
18+
## Possible causes
19+
20+
- Unstructured tried to save its processed data to the specific Amazon S3 bucket, but Unstructured could not find the necessary
21+
security keys or other credentials to get access to that bucket.
22+
- The specified AWS Access Key ID and Secret Access Key are missing from your S3 destination connector configuration.
23+
- The specified credentials are incorrect, have been revoked, or have expired.
24+
- The S3 destination connector is configured to use a temporary IAM role, but that role does not have permission to write to the target bucket.
25+
26+
## Possible solutions
27+
28+
- Double-check that your workflow's Amazon S3 destination connector has the correct, valid credentials specified.
29+
- Confirm that the specified credentials are still present and valid in your AWS account.
30+
- If your organization uses IAM roles, contact your platform administrator to ensure the role has `s3:PutObject` permissions for the bucket you are trying to use.
31+
32+
## Additional resources
33+
34+
- Always copy and paste security keys and other credentials carefully.
35+
- If possible, use your organization's secrets manager to handle security keys and other credentials automatically.
36+
- To ask questions or get additional help with this issue, see [requesting support](/support/request).

‎support/issues/get-authenticated-user-error.mdx‎

Lines changed: 32 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,32 @@
1+
---
2+
title: Get authenticated user error
3+
---
4+
5+
## Issue
6+
7+
When you try to sign in to Unstructured, you get an error that contains text similar to the following:
8+
9+
```text
10+
Error in getAuthenticatedUser
11+
```
12+
13+
This error can prevent you from accessing your
14+
Unstructured account dashboard or performing other actions in your Unstructured account.
15+
16+
## Possible causes
17+
18+
- Unstructured is having trouble verifying your login session.
19+
- Your Unstructured login session has expired.
20+
- Your web browser's cache or cookies are causing some kind of conflict.
21+
- There is a temporary issue with Unstructured's authentication service.
22+
23+
## Possible solutions
24+
25+
- Log out of your Unstructured account completely, if you can. To do this, in the Unstructured user interface, on the sidebar,
26+
click your user icon, and then click **Logout**.
27+
- Clear your web browser's cache and cookies for the Unstructured website.
28+
- Close and then reopen your web browser, and then try to sign in again to your Unstructured account.
29+
30+
## Additional resources
31+
32+
To ask questions or get additional help with this issue, see [requesting support](/support/request).

‎support/issues/google-drive-schema-validation.mdx‎

Lines changed: 52 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,52 @@
1+
---
2+
title: Google Drive schema validation failure
3+
---
4+
5+
## Issue
6+
7+
When you save and test a Google Drive source connector, you get an error that contains text similar to the following:
8+
9+
```text
10+
Schema validation failure for GoogleDriveConnectionConfig:
11+
Value error, Expecting value: line 1 column 1 (char 0) at
12+
access_config,service_account_key got -----BEGIN PRIVATE KEY-----
13+
...
14+
```
15+
16+
Although you can finish creating and saving this connector, the connection test fails, and you cannot successfully run any
17+
workflows that use this failed connector.
18+
19+
## Cause
20+
21+
You have configured the Google Drive source connector with the Google Drive service account's key in an incorrect format.
22+
You likely have specified only a portion of the file, such as just the private key part. Unstructured expects the entire contents
23+
of the Google Drive service account's key file.
24+
25+
## Solution
26+
27+
1. Find the original Google Drive service account's key file in JSON format, which you downloaded earlier from your Google Cloud account when
28+
you originally created the Google Drive source connector.
29+
2. Open the file in a text editor such as Notepad, Visual Studio Code, or TextEdit.
30+
3. Copy everything in the file, starting from the first opening brace `{` to the final closing brace `}`.
31+
4. In the original Google Drive source connector's settings, delete the original contents of the **Service Account Key** field, and paste
32+
the complete JSON content that you just copied.
33+
5. Save the changes and re-test and re-save the connector.
34+
35+
## Additional resources
36+
37+
- If the connector test still fails, try using a tool such as `tr` for macOS or Linux, or PowerShell for Windows, to copy
38+
the entire contents of the file, instead of a text editor, as follows:
39+
40+
- For macOS or Linux:
41+
42+
```text
43+
tr -d '\n' < <path-to-downloaded-key-file>
44+
```
45+
46+
- For Windows:
47+
48+
```text
49+
(Get-Content -Path "<path-to-downloaded-key-file>" -Raw).Replace("`r`n", "").Replace("`n", "")
50+
```
51+
52+
- To ask questions or get additional help with this issue, see [requesting support](/support/request).

‎support/issues/no-fast-partitioning-for-images.mdx‎

Lines changed: 33 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,33 @@
1+
---
2+
title: The Fast strategy is not available for image files
3+
---
4+
5+
## Issue
6+
7+
When you try to run a workflow in Unstructured with the following settings:
8+
9+
- The workflow contains a **Partitioner** node set to use the **Fast** partitioning strategy.
10+
- The workflow is configured to process an image file, such as a PNG or JPG file.
11+
12+
You get an error that contains text similar to the following, and the workflow fails to run:
13+
14+
```text
15+
422: Error in partition - {"detail":"The fast strategy is not available for image files"}
16+
```
17+
18+
## Cause
19+
20+
Unstructured is trying to process an image file&mdash;such as a PNG or JPG file&mdash;by using the **Fast** partitioning strategy, but
21+
this strategy is designed for text-based documents only and cannot analyze the layout of images.
22+
23+
## Solution
24+
25+
1. Change the workflow's **Partitioner** node's **Partition Strategy** setting to **Auto**, **High Res**, or **VLM**.
26+
2. Re-run the workflow.
27+
28+
## Additional resources
29+
30+
- If your workflow is configured to process a mixture of file types, you should use the **Auto** partitioning strategy. Unstructured will then
31+
automatically choose the best partitioning strategy for each file.
32+
- Learn more about the various [partitioning](/ui/partitioning) strategies.
33+
- To ask questions or get additional help with this issue, see [requesting support](/support/request).

‎support/issues/workflow-job-in-progress.mdx‎

Lines changed: 28 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,28 @@
1+
---
2+
title: Workflow already has a job scheduled or in progress
3+
---
4+
5+
## Issue
6+
7+
When you try to run a workflow, you get a warning message that contains text similar to the following:
8+
9+
```text
10+
Failed to schedule job for workflow ... Workflow already has a job scheduled or in progress. Retrying in 60 seconds.
11+
```
12+
13+
## Causes
14+
15+
- Unstructured tried to start a new job run for the workflow, but the previous job run for that workflow is stuck or has otherwise not finished yet.
16+
- The workflow's schedule is too frequent for how long the job takes to run. For example, the workflow might be scheduled to
17+
run every 5 minutes, but each job run for that workflow typically takes 10 minutes or longer to complete.
18+
19+
## Solutions
20+
21+
- Unstructured will wait and try the new job run again in 60 seconds.
22+
- Check previous job runs for the workflow to see if any of them are stuck. If so, try to cancel them.
23+
- Check previous job runs for the workflow to see if any of them seem to be running much longer than usual. If so,
24+
consider adjusting the workflow's schedule to run less frequently. Or, temporarily turn off the workflow's schedule and then run the workflow manually as needed.
25+
26+
## Additional resources
27+
28+
To ask questions or get additional help with this issue, see [requesting support](/support/request).

0 commit comments

Comments
 (0)