Incident Missing Entities
Good morning! I would like to have some clarification on how entities work. Yesterday I found out that if I have 2 entities of the same type (In this particular case, two entities of the type Account), with the same identifier (originally, both share the identifier 'Name'), Sentinel appears to throw away one of them, or both in some instances, and when the alert generates an incident, the entities defined won't appear. I have switched out the identifier on both account types to something different, but until an incident gets triggered, I can't confirm if this will fix the original issue. So my questions are An analytic rule can or can't have two entities of the same type defined? If yes, that means that they need to have different identifiers. Is this a correct asumption? Some identifiers expect a certain type of value to be assigned, that means that eventually, I can ran out of identifiers for my entities or face the added complexity of dealing with types when returning values from my KQL query What could happen if I map an identifier to something that matches the type but not what that identifier represents? in this case, for Account, we have the identifier ObjectGuid. If I assign a value type string to it, that is not a guid, wouldn't that mess up something else in the background? Example, incident grouping If I move instead to use Sentinel Entities, which appear to be the 'general' option, I could only use one, since I only have 'entity' as available identifier, looping back to the problem of can have only one type of identifier for identity type. Thanks in advance
IdentityInfo with analytics KQL query
Hi, I'm currently trying to create a KQL query for an alert rule in Sentinel. The log source upon which the alert rule is based, only contains the SAMAccountName, which prevents me from mapping it to an Account entity in the alert. I'm therefore trying to use the IdentityInfo table to lookup the AadUserId of the user, using the SAMAccountName. The issue I'm running into is that I want my query to run every 10 minutes, and look up data from the past 10 minutes, as this is most suitable given the nature of the alert and the log source. This however causes the lookup in the IdentityInfo table to also only check data from the last 10 minutes, which doesn't work as the data in that table may be much older and therefor fail the lookup of the AadUserId of the user. According to the documentation, the IdentityInfo table is refreshed every 14 days, so for it to work I'd have to create a query that checks all logging, including that of the log source, from the past 14 days, which is not what I want. Hopefully some of you have suggestions or ideas on how to make this work. Thanks a lot! MarekPlaybook when incident trigger is not working
Hi I want to create a playbook to automatically revoke session user when incident with specifics title or gravity is created. But after some test the playbook is'nt run autimacally, it work when I run it manually. I did'nt find what I do wrong. See the image and the code bellow. Thanks in advance! { "definition": { "$schema": "https://schema.management.azure.com/providers/Microsoft.Logic/schemas/2016-06-01/workflowdefinition.json#", "contentVersion": "1.0.0.0", "triggers": { "Microsoft_Sentinel_incident": { "type": "ApiConnectionWebhook", "inputs": { "host": { "connection": { "name": "@parameters('$connections')['azuresentinel']['connectionId']" } }, "body": { "callback_url": "@{listCallbackUrl()}" }, "path": "/incident-creation" } } }, "actions": { "Get_incident": { "type": "ApiConnection", "inputs": { "host": { "connection": { "name": "@parameters('$connections')['azuresentinel-1']['connectionId']" } }, "method": "post", "body": { "incidentArmId": "@triggerBody()?['object']?['id']" }, "path": "/Incidents" }, "runAfter": {} }, "Send_e-mail_(V2)": { "type": "ApiConnection", "inputs": { "host": { "connection": { "name": "@parameters('$connections')['office365']['connectionId']" } }, "method": "post", "body": { "To": "email address removed for privacy reasons", "Subject": "Ceci est un test", "Body": "</p> <p class="\"editor-paragraph\"">@{body('Get_incident')?['id']}</p> <p class="\"editor-paragraph\"">@{body('Get_incident')?['properties']?['description']}</p> <p class="\"editor-paragraph\"">@{body('Get_incident')?['properties']?['incidentNumber']}</p> <p>", "Importance": "Normal" }, "path": "/v2/Mail" }, "runAfter": { "Get_incident": [ "Succeeded" ] } } }, "outputs": {}, "parameters": { "$connections": { "type": "Object", "defaultValue": {} } } }, "parameters": { "$connections": { "type": "Object", "value": { "azuresentinel": { "id": "/subscriptions/xxxx/providers/Microsoft.Web/locations/xxxxx/managedApis/xxxxxxx", "connectionId": "/subscriptions/xxxxxxx/resourceGroups/xxxxxx/providers/Microsoft.Web/connections/azuresentinel-Revoke-RiskySessions1", "connectionName": "azuresentinel-Revoke-RiskySessions1", "connectionProperties": { "authentication": { "type": "ManagedServiceIdentity" } } }, "azuresentinel-1": { "id": "/subscriptions/xxxxxx/providers/Microsoft.Web/locations/xxxx/managedApis/xxx", "connectionId": "/subscriptions/xxxxxxx/resourceGroups/xxxxx/providers/Microsoft.Web/connections/xxxx", "connectionName": "xxxxxx", "connectionProperties": { "authentication": { "type": "ManagedServiceIdentity" } } }, "office365": { "id": "/subscriptions/xxxxxx/providers/Microsoft.Web/locations/xxxxx/managedApis/office365", "connectionId": "/subscriptions/xxxxx/resourceGroups/xxxxxx/providers/Microsoft.Web/connections/o365-Test_Send-email-incident-to-xxxx", "connectionName": "o365-Test_Send-email-incident-to-xxxxx" } } } } }Unknown Behaviour Involving GroupsService in OfficeActivity
I have spotted a few hundred events with the following KQL query in my environment. This is the result of one of the entries. It looks like a regular legitimate behaviour by Microsoft but I don't seem to find any documentation about it. Can anyone share the insight of it? Thank you!Lookup data from the last == ingestion_time()
Howdy! In "Analytics rule wizard - Create a new Scheduled rule" under Query scheduling you have to fill out "Lookup data from the last" What time field is Sentinel looking at when determine which events to include in the lookup data? Is it ingestion_time()? Is it TimeGenerated? How does it know?
Daily "Network Port Sweep detected on port x" but no Source IP
For a couple of months we have been getting "Network Port Sweep was detection by multiple IPs" with ports 135 and 445 mostly. The KQL attached lists a load of Destination IPs but no Source IP (see example below) All these are inside our network, and so not coming from outside the network where the firewall would block such scans. We suspect it is one of our network tools such as Solarwinds but without the Source IP it is difficult to prove anything. // The query_now parameter represents the time (in UTC) at which the scheduled analytics rule ran to produce this alert. set query_now = datetime(2024-06-24T03:01:10.2587643Z); let lookback = 1h; let threshold = 20; _Im_NetworkSession(starttime=ago(lookback), endtime=now()) | where NetworkDirection == "Inbound" | summarize make_set(DstIpAddr, 100) by DstPortNumber | where array_length(set_DstIpAddr) > threshold Tried to modify the query, this is the KQL I have tried. set query_now = datetime(2024-01-26T07:47:48.3756000Z); let lookback = 1h; let threshold = 20; _Im_NetworkSession(starttime=ago(lookback), endtime=now()) | where NetworkDirection == "Inbound" | summarize make_set(DstIpAddr, 100), make_set(SrcIpAddr, 100) by DstPortNumber | where array_length(set_DstIpAddr) > threshold Anyone got any suggestions on how we can track down what is causing this alert in Sentinel?
Anomaly Excessive NXDOMAIN DNS Queries - analytics rule
I have noticed that we see quite a few endpoints that are triggering the Excessive NXDOMAIN DNS Queries anomaly analytics rule in Microsoft Sentinel. When I investigate these for tuning purposes, I see that the vast majority of these queries (in the in-addr.arpa domain) are for IP addresses owned by Microsoft. It appears that Microsoft have no interest in publishing reverse DNS entries, because I am unable to resolve them from any online DNS tools. The whois records do point to Microsoft, though. What's a good way to either stop this from happening, or eliminate the Microsoft IP address space from the query results?
