The instance now seems to be succeeding on segmentation model imports and failing on transcription model imports. Trying to import a transcription model consistently leads to a screen like the following:
The requested URL was rejected. Please consult with your administrator.
Your support ID is: 13895239713375090222
[Go Back]
Googling the above error message suggests that that particular message tends to come up when a firewall is blocking something.
This problem seems to have been ongoing since Oct 17th at the latest, since the data from that round of import tests shows that the transcription model uploads never completed. The issue therefore predates changes to the VM configurations.
Ran another round of testing today using the following four transcription models, all of which failed:
Meanwhile, none of the segmentation models I have tried to upload through the GUI have failed. However, using the API to try to upload models is now failing regardless of whether they are segmentation or transcription models.
Tests so far
It does not matter what user account I try to use to upload the transcription models.
It does not make a difference whether I am connected directly to the PU network or using the VPN for remote access.
I have tested both Firefox and Chrome with no luck.
Trying to upload via the API fails on both segmentation and transcription models, with the same error message.
Trying to force an upload as admin through the admin page leads to the same error message.
Lying to eScriptorium in the admin page and claiming that the transcription model file is actually a segmentation model still leads to the same error message.
Conversely, lying to eScriptorium in the admin page and claiming that a segmentation model file is actually a transcription model leads to a successful upload.
A sampling of the support IDs I've been getting in the error messages
Describe the bug
The instance now seems to be succeeding on segmentation model imports and failing on transcription model imports. Trying to import a transcription model consistently leads to a screen like the following:
Googling the above error message suggests that that particular message tends to come up when a firewall is blocking something.
This problem seems to have been ongoing since Oct 17th at the latest, since the data from that round of import tests shows that the transcription model uploads never completed. The issue therefore predates changes to the VM configurations.
Ran another round of testing today using the following four transcription models, all of which failed:
Meanwhile, none of the segmentation models I have tried to upload through the GUI have failed. However, using the API to try to upload models is now failing regardless of whether they are segmentation or transcription models.
Tests so far
A sampling of the support IDs I've been getting in the error messages