Issue with audit 3.4

[andersonm]

Hi all,

Excuse me for my English.

I'm trying to use the new auditing 3.4. I want to persist the audit always that an user consult a document.
I'm able to do this, but not all consultations, only the first consult in each document. E.g., an user creates a document and download it, then it is audited, but if either the same or another user download the same document again, it isn't audited.

So far, I did this:

1  Enabled the audit in the alfresco-global.properties:

# Enable audit in general
audit.enabled=true

# Enable the alfresco-access audit application
audit.alfresco-access.enabled=true

# Enable the auditing of sub-actions. Normally disabled as these values are
# not normally needed by audit configurations, but may be useful to
# developers
audit.alfresco-access.sub-actions.enabled=true‍‍‍‍‍‍‍‍‍‍

2 Created a file alfresco-audit-document.xml, placed it in <TOMCAT_HOME>/shared/classes/alfresco/extension/audit, and populated it with the follow code:

<?xml version='1.0' encoding='UTF-8'?>

    <!– Node Functions –>

    <Audit
        xmlns="http://www.alfresco.org/repo/audit/model/3.2"
        xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
        xsi:schemaLocation="http://www.alfresco.org/repo/audit/model/3.2 alfresco-audit-3.2.xsd"
        >

        <DataExtractors>
      <DataExtractor name="simpleValue" registeredName="auditModel.extractor.simpleValue"/>
      <DataExtractor name="nodeName" registeredName="auditModel.extractor.nodeName"/>
      <DataExtractor name="nodeType" registeredName="auditModel.extractor.nodeType"/>
        </DataExtractors>

        <DataGenerators>
           <DataGenerator name="personFullName" registeredName="auditModel.generator.personFullName"/>
           <DataGenerator name="currentUser" registeredName="auditModel.generator.user"/>
           <DataGenerator name="systemTime" registeredName="auditModel.generator.time"/>
        </DataGenerators>

        <PathMappings>
            <PathMap source="/alfresco-api/post/ContentService/getReader" target="/document/read"/>
        </PathMappings>

        <Application name="AuditDocument" key="document">           
            <AuditPath key="read">
                <AuditPath key="no-error">
                    <RecordValue key="document" dataExtractor="nodeName" dataSource="/document/read/args/nodeRef"/>
                    <GenerateValue key="user" dataGenerator="personFullName"/>
                </AuditPath>
                <AuditPath key="error">
                    <RecordValue key="document" dataExtractor="nodeName" dataSource="/document/read/args/nodeRef"/>
                    <GenerateValue key="user" dataGenerator="personFullName"/>
                </AuditPath>
            </AuditPath>
        </Application>
</Audit>‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

With this, I use the command "curl -u admin:admin "http://localhost:8080/alfresco/service/api/audit/query/AuditDocument?verbose=true"" and get a result like this:

{
         "id":2605,
         "application":AuditDocument,
         "user":admin,
         "time":"2011-08-24T12:18:58.284-03:00",
         "values":
         {
                     "\/document\/read\/no-error\/document":"111001-ADD-001-A00.js"
                     ,"\/document\/read\/no-error\/user":"Administrator"
         }
        
      },

It's ok, but have the problem described above.

I'm using alfresco commmunity 3.4.

Hope I could be clear.

Best regards.

[jneeve]

I have the same problem and am waiting to hear back from support if this is a bug, or some sort of desired behaviour (?!)

In the old audit config (>= 3.3) sometimes multiple reads done quickly by the same user would trigger only one audit, but if multiple users read the content there would always an audit read per user.

[andersonm]

jneeve

Thanks for your reply.

Any news plz post here.

[jneeve]

Looks like its a problem with browser caching…. Try re-doing your test with diferent browswers.

I've retested using IE and Chrome, and you will get a unique audit the first time. But depending on your browser cache settings, the second download attempt won't actualy trigger a read again since the file comes from the cache…

Personally, I think each request to read should be audited regardless, but thats the way it works.