Feed Files
Integrations > Data Import > Imported Files > Feed Files
The Feed Files screen tracks, manages, and reports on each of the most recent imported feed-files. Feed files contain the raw information that creates and updates surveys, survey invitations, and company-specific details including Units and unit groups.
Use this screen to:
- Confirm that a feed-file was delivered to Medallia Experience Cloud, and to get information on the current status of that file.
- See how this feed file entered Experience Cloud (such as via SFTP/FTP push or pull, or email import).
- Examine feed files.
- Reprocess a feed file, if necessary.
The list includes the feed files from the last 365 days. Longer periods of time are available via Medallia Support and requesting the range to display.
Using the Feed Files screen
After a file has been imported and processed, it is listed on the left side of the screen, with the most recent at the top and the oldest at the bottom of the list.
- Select a file to see the processing results and to perform further processing actions on the file.
- Filter the results by selecting the Auto Importer Processors that processed the feed file.
- The filenames are color-coded to provide a quick status:
- Black — The file was processed successfully in active mode.
- Purple — There was a file processing error.
- Yellow — The file was processed successfully in pretend mode.
- Grey — The file was preprocessed and is ready to be processed.
The right side of the screen lists the properties and provides action buttons for the selected file.
Feed file ID
Every feed-file has a unique ID, and every record created from a feed-file records that ID in the a_feed_file_id A-Field. To locate a feed-file by its ID, in the Search box of the file list, enter the ID, or partial ID to see all files that include that part of the sequence.
Properties
- File
-
- File Name
- Name of the processed file, the same name as in the list of processed files. If the file was encrypted (such as
.pgp) or in an archive (such as a .zip file), this is the unencrypted actual feed file.Click Download to download and save the file locally.
Note: This is the file as it came into Experience Cloud. If the file was encrypted, then this is also encrypted: you cannot read the contents of an encrypted file. Re-selecting the filename usually corrects this condition. - File Size
- Size of the file in Bytes.
- Position in Queue
- The approximate position of the file in the processing queue.
- Number of Files in Queue
- The approximate total number of files in the processing queue.
- Created
- Timestamp when the file was fed into Experience Cloud.
- Modified
- Timestamp when the file was last modified.
- Feed Account
-
- Type
- Type of Feed Account that imported the file, such as SFTP, FTP, Email, Web, or Feed pull. When the file was uploaded from the Auto Importer Processors screen, the Type is Setup Upload.
- Name
-
Name of the Feed Account that received or pulled the file as it was when it entered the system. Typically this is the same name as the File Name property; however, when the file is encrypted, then this filename extension is
.pgp.For manually uploaded file, the name indicates who uploaded the file and when, such as "Upload by Skyler on 2016-05-26 16:51:10".
- Record Processor
-
- State (when last processed)
-
State of the Feed Account when the feed file was processed, as set by the Processing mode property for the feed:
- Active — The file was processed in active mode.
- Upload with pretend processing — The file was processed in pretend mode.
- Name
- Name of the Auto Importer specification that processed the file.
- Processing
-
- Processing history
-
List of the processing activities performed on the file, with the most recent at the top of the list.
Date
Timestamp when the activity happened.
Status
The status of the activity, including:
- Processed — The feed file processed without error; however, you must still check the processing report to confirm that all went as expected.
- Waiting — Processing is paused while waiting for the next action. Files to be processed are placed in a queue and are in Waiting state until the processing begins. Files can also be in this state after processing a file in pretend-mode. In this state you can click Cancel processing and Reprocess file.
-
Decompression failed — Could not decompress the file. A compressed file has one of these filename extensions:
gz,bz2,z,zip, ortar. - Decryption failed — The file appears to be encrypted, but it is not a Open PGP ASCII Armored message.
-
File processing was canceled — Someone manually cancelled the process.
- No processing — No Record processor was selected, or this feed file is the end result of a Preprocessor test. No further processing is expected on it.
- Processing error — Some other error occurred. See the Description for details.
- Signature verification failed — There is something wrong with the GPG or PGP file signature.
- Unknown file format — There was no data file in the Decompressed feed file.
- Unknown key used to encrypt the file — The secret key for the GPG or PGP file is not available.
- XML parsing failed — The file is supposed to be XML, but it is not proper XML.
-
XSLT file transform error — The system encountered a problem processing the importer specification.
-
Null — Web Feeds in Synchronous web feed mode do not generate normal processing reports or processing records. Instead, they generate periodic processing report records for the files that have been processed. These records can encompass more than one file and they do not have status. Note that, when a file processing fails in this scenario, it generates an imported file with a non-null status value.
Description
Detailed description of the Status. When the Status indicates an error occurred, this is a detailed description of the error. The description matches the error(s) listed in the processing report. For preprocessor results, the description includes information on the preprocess action performed in the step.
Details
Links to reports with more detailed information. See Import Processing Reports for details about the reports.
- Last processed batch
-
The count of the last processed batch of records.This property appears after processing at least one batch of records in active mode.
In active mode, the importer processes feed files in batches of about 5,000 records. When there are more than that many records in the file, this property reports how many of the batches have been processed. For example, when there are about 10,000 records in the file, the last processed batch will be 2 or 3. Batch processing does not happen in pretend mode.
When the processing of the feed file is incomplete and has stopped, clicking Reprocess on Active restarts processing with the next batch of records – the batch it was processing when it was stopped. So, for example, if there were 20,000 records, and the last process batch was 2, processing will restart with the third batch because that is the one that was in process when interrupted (usually record number 10,001).
The Reset last processes batch button resets the batch-count to zero (0) such that when reprocessing restarts, the entire file is reprocessed from the beginning of the file instead of the next batch.
- Cancel Processing
-
Cancels the processing of this feed file. Works both for files that are in the Waiting status (in queue to be processed), as well as files that are currently being processed. If it is currently being processed, a best-effort attempt to terminate the processing task.
Canceling a process also cancels the processing report. As such, it is best to use this in pretend mode only.
- Reprocess feed file
-
Reprocess the feed file from the beginning of the file, and use the mode specified in the Processing Mode property of the Feed specification.
This is useful for debugging a feed file by processing it repeatedly in pretend mode, and reviewing the processing report to ensure skipped count and number of exceptions meet your expectation.
When ready to process the file for real and add the records to the database, use Reprocess on Active, or change the Processing Mode to active.
- Reprocess on Active
-
Reprocess the feed file in active-mode, regardless of the Processing Mode property of the Feed specification.
Reprocessing the file starts with the beginning of the file unless the Last processed batch indicates some records have already been processed in active mode. In that case, reprocessing starts with the next batch of records. See the description of that property for details.
To reprocess multiple failed files at one time, use the Bulk Retry screen.
- Generate Excel Report
- Generates an Excel processing report for the most recent time the Feed File was processed. This works only when there is a Processing Report Data report listed in the Details column of the Processing History table. After clicking Generate Excel Report, you must refresh your browser for the new report to appear.
In the Details column, the on-demand Excel report is named Detailed Async Excel Report. If Experience Cloud encounters an error when trying to generate the report, it produces a Detailed Sync Excel Report based on the most recently-processed data. Both the Async and Sync versions of the report appear the same. For more information about the data and formatting of the Excel report, Import report (Excel file).
Restriction: By default, all detailed processing reports for concurrent asynchronous Web feeds older than a maximum of 90 days are no longer available; data is swept and is permanently purged after that period. Contact your Medallia expert to understand the configured limits for your instance
- Other actions
-
- Reset Last Modified
- Sets the last-modified timestamp to 1 January 1970.
- Delete non-sampled episodes
- Removes non-sampled episode records from the pool when using pool-based sampling. This is useful, for example, to remove records with incorrect information. The screen shows the count of records removed when complete.
- Count Episodes
-
Reports the count of episodes records sampled and not sampled (such as those blocked by filters and quarantine). For example, a report might say "3 episodes (2 samples, 1 non-sampled).
Removing non-sampled records from the pool (via Delete non-sampled episodes) changes the # non-sampled count to zero (0).
