How to Specify What Information is Included in Log Files

To specify what information is included in log files, use the following graphic and procedure.

1. In the AutoClassifier administration portal, click Logging from the left navigation panel.

2. Configure the following fields:

   1. Capture Pipeline Statistics:

      1. Pipeline Statistics capture performance indicators for all components in the AutoClassifier Pipeline when enabled (True)

      2. Default: True

   2. Enable Rules Engine Document Trace:

      1. When enabled (True) the document trace feature captures detailed analysis of the Rules Engine Component.

      2. Default: False

   3. Include All Queries in Document Trace:

      1. By default the Document Trace will only include potential Rule matches.

      2. If set to True, all queries from the taxonomy will be included

      3. Default: False

   4. Document Property Trace Filter:

      1. This filter is property-based.

         1. For example: filename=”Test AC Document.docx”

      2. Any file with metadata that matches the trace is included when the trace is enabled

      3. However, no filter produces a trace for every profiled document

      4. If you do not want all of your document properties included in the logging operation, specify the list of properties to include using a comma (,) separated list

      5. This information is stored in the DocumentTrace folder

   5. Logging Servers:

      1. Each installation of the AutoClassifier Engine automatically registers the server for logging

      2. This is based upon the initial IIS Bindings

      3. If the binding changes, each server needs to be updated to reflect the new bindings

3. Log Level:

   1. Select the Log Level to set on all servers

4. Download Logs:

   1. Retrieves all accessible logs for the current day and merges the data into a single file

5. Clear Logs:

   1. Clears all accessible log files on all servers

6. Click the Save button at the bottom of the page to save your changes.

Scheduled Annotation: Using Logs to Monitor Library Processing

Scheduled annotation can be useful for monitoring progress when annotating document libraries.

To monitor Scheduled Annotation progress, use the logging feature.

To get visibility into the progress of a library being processed, use the log file results, as described and shown below.

During scheduled annotation all of the following are annotated (if so configured):

• Sites

• Site collections

• (Document) Libraries

For each document library, the following data is shown:

• Report items processed

• List item IDs

  • Numbers ending encircled in screenshot below, ending in _1997, _1996, _1995, _1994, etc.

  • Example: 076786DA-BF40-44D4-8F37-0DCF57A3488A_1996

  • Format: ListID_ItemID

• Result of batch processing:

  • The keys shown in the screenshot for batch processing below are a series of numbers which correlate to: site collection ID, site ID, and library ID

    • Use the key to troubleshoot issues, if required.

    • Format: SiteCollectionID_SiteID_LibraryID

• DPS

  • Documents per second (for each batch of 20 items, sorted by list key)

• Progress so far

  • Percentage of completion for each library the job is processing

  • Note:

    • The progress percentage is not perfectly accurate in all scenarios.

    • Custom filtering, such as CAML queries, used to filter library items in SharePoint, the percentage is not accurate.

    • Also, if the job failed, and continued from where it left, the percentage is “pessimistic” meaning it will not consider items previously processed and will show a lower value than the real one.

    • When you are not using custom CAML queries, and your job works uninterrupted, the percentage is perfectly accurate.

• All Items updated, including those with metadata

  • To count total number of items updated, it must be done at the Debug log level:

    • Succesfully annoated item: "ItemTitle"