DDUs for Log Monitoring

This page explains the concept of Log Monitoring, how we calculate DDUs for Log Monitoring, and how you can estimate and track DDU consumption if you use Log Monitoring.

What is Log Monitoring?

With Log Monitoring as a part of the Dynatrace platform, you gain direct access to the log content of all your mission-critical processes. You can create custom log metrics for smarter and faster troubleshooting. You will be able to understand log data in the context of your full stack, including real user impacts.

How does Dynatrace calculate DDU consumption for Log Monitoring?

The Davis data units (DDUs) model counts all incoming log records (entries) from your log data. Each log record (line, message, entry) deducts 0.0005 DDU from your available quota. For example, 1 million log records multiplied by a DDU weight of 0.0005 consumes a total of 500 DDUs.

A log record is recognized in two ways:

Timestamp
JSON object

Timestamp

Dynatrace assumes that a timestamp is the beginning of a new log record.

For example, in the following log data (consumed via log file or generic ingestion), Dynatrace counts nine log records based on timestamp occurrence:

Log file input

plaintext
1: Oct 18 05:56:11 INFO ip-10-176-34-132 DHCPREQUEST on eth0 to 10.176.34.1
2: Oct 18 05:56:12 INFO ip-10-176-34-132 DHCPACK from 10.176.34.1
3: Oct 18 05:56:13 INFO ip-10-176-34-132 bound to 10.176.34.132 -- renewal in 1551s4:
4: Oct 18 05:56:15 INFO ip-10-176-34-132 [get_meta] Getting token for IMDSv
5: Oct 18 05:56:16 INFO ip-10-176-34-132 [get_meta] Trying to get http://169.23.2.3
6: Oct 18 05:56:18 INFO ip-10-176-34-132 [rewrite_aliases] Rewriting aliases
7: Oct 18 06:22:06 INFO ip-10-176-34-132 DHCPREQUEST on eth0 to 10.176.34.1 port 67
8: Oct 18 06:22:07 INFO ip-10-176-34-132 DHCPACK from 10.176.34.1 (xid=0x3a182c8c)
9: Oct 18 06:22:10 INFO ip-10-176-34-132 bound to 10.176.34.132 -- renewal in 1364s

Generic ingestion input

shell
curl -X POST "https://my.activegate/api/v2/logs/ingest"
     -H  "accept: application/json; charset=utf-8"
     -H  "Content-Type: text/plain; charset=utf-8"
     -d "Oct 18 05:56:11 INFO ip-10-176-34-132 DHCPREQUEST on eth0 to 10.176.34.1Oct 18 05:56:12 INFO ip-10-176-34-132 DHCPACK from 10.176.34.1Oct 18 05:56:13 INFO ip-10-176-34-132 bound to 10.176.34.132 -- renewal in 1551s4:Oct 18 05:56:15 INFO ip-10-176-34-132 [get_meta] Getting token for IMDSvOct 18 05:56:16 INFO ip-10-176-34-132 [get_meta] Trying to get http://169.23.2.3Oct 18 05:56:18 INFO ip-10-176-34-132 [rewrite_aliases] Rewriting aliasesOct 18 06:22:06 INFO ip-10-176-34-132 DHCPREQUEST on eth0 to 10.176.34.1 port 67Oct 18 06:22:07 INFO ip-10-176-34-132 DHCPACK from 10.176.34.1 (xid=0x3a182c8c)Oct 18 06:22:10 INFO ip-10-176-34-132 bound to 10.176.34.132 -- renewal in 1364s"

The DDU cost is 9 log lines/records x 0.0005 log weight = 0.0045 DDUs.

JSON object

Dynatrace assumes that a single JSON object is a log record. A JSON file can contain multiple objects that count as a log record.

For example, in the following log data, Dynatrace counts three log records based on JSON object occurrence:

Log file

json
[
  {
  "timestamp": "2021-07-29T10:54:40.962165022Z",
  "level": "error",
  "log.source": "/var/log/syslog",
  "application.id": "PaymentService-Prod",
  "content": "DHCPREQUEST on eth0 to 10.176.34.1"
  },
  {
  "log.source": "/var/log/syslog",
  "content": "[get_meta] Getting token for IMDSv"
  },
  {
  "content": "DHCPACK from 10.176.34.1 (xid=0x3a182c8c)"
  }
]

Generic Ingestion

shell
curl -X POST "https://my.activegate/api/v2/logs/ingest"
     -H  "accept: application/json; charset=utf-8"
     -H  "Content-Type: application/json; charset=utf-8"
     -d "[{\"timestamp\":\"2021-07-29T10:54:40.962165022Z\",\"level\":\"error\",\"log.source\":\"/var/log/syslog\",\"application.id\":\"PaymentService-Prod\",\"content\":\"DHCPREQUEST on eth0 to 10.176.34.1\"},{\"log.source\":\"/var/log/syslog\",\"content\":\"[get_meta] Getting token for IMDSv\"},{\"content\":\"DHCPACK from 10.176.34.1 (xid=0x3a182c8c)\"}]"

The DDU cost is 3 log lines/records x 0.0005 log weight = 0.0015 DDUs

How Log Monitoring can affect your DDU consumption

Dynatrace counts log records based on timestamp, even if the log record contains an attached stack trace.

For example, both of the following logs (Log1 and Log2) contain 14 lines of log data. Dynatrace calculates log records that consume DDUs on valid timestamp. As a result, Log1 generated cost of 0.007 DDUs while Log2 generated a cost of 0.001 DDUs.

Log1

plaintext
1:  Oct 18 05:56:11 INFO ip-10-176-34-132 DHCPREQUEST on eth0 to 10.176.34.1
2:  Oct 18 05:56:12 INFO ip-10-176-34-132 DHCPACK from 10.176.34.1
3:  Oct 18 05:56:13 INFO ip-10-176-34-132 bound to 10.176.34.132 -- renewal in 1551s4:
4:  Oct 18 05:56:15 INFO ip-10-176-34-132 [get_meta] Getting token for IMDSv
5:  Oct 18 05:56:16 INFO ip-10-176-34-132 [get_meta] Trying to get http://169.23.2.3
6:  Oct 18 05:56:15 INFO ip-10-176-34-132 [get_meta] Getting token for IMDSv
7:  Oct 18 06:16:16 INFO ip-10-176-34-132 [get_meta] Trying to get http://169.23.2.3
8:  Oct 18 06:16:18 INFO ip-10-176-34-132 [rewrite_aliases] Rewriting aliases
9:  Oct 18 06:21:26 INFO ip-10-176-34-132 DHCPREQUEST on eth0 to 10.176.34.1 port 67
10: Oct 18 06:22:06 INFO ip-10-176-34-132 DHCPREQUEST on eth0 to 10.176.34.1 port 67
11: Oct 18 06:22:07 INFO ip-10-176-34-132 DHCPACK from 10.176.34.1 (xid=0x3a182c8c)
12: Oct 18 06:22:10 INFO ip-10-176-34-132 bound to 10.176.34.132 -- renewal in 1364s
13: Oct 18 14:51:22 ERROR ip-10-176-34-32 classOne: Index out of range
14: Oct 18 16:52:10 ERROR ip-10-176-34-32 classOne: Index out of range

The DDU cost is 14 log records x 0.0005 log weight = 0.007 DDUs.

Log2

plaintext
1: Oct 18 14:51:22 ERROR ip-10-176-34-32 classOne: Index out of range
2:     java.lang.StringIndexOutOfBoundsException: String index out of range: 18
3:     at java.lang.String.charAt(String.java:658)
4:     at com.example.app.loggingApp.classOne.getResult(classOne.java:15)
5:     at com.example.app.loggingApp.AppController.tester(AppController.java:27)
6:     at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
7:     at sun.reflect.NativeMethodAccessorImpl.invoke0(NativeMethodAccessorImpl.java:62)
8: Oct 18 16:52:10 ERROR ip-10-176-34-32 classOne: Index out of range
9:     java.lang.StringIndexOutOfBoundsException: String index out of range: 18
10:    at java.lang.String.charAt(String.java:658)
11:    at com.example.app.loggingApp.classOne.getResult(classOne.java:15)
12:    at com.example.app.loggingApp.AppController.tester(AppController.java:27)
13:    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
14:    at sun.reflect.NativeMethodAccessorImpl.invoke0(NativeMethodAccessorImpl.java:62)

The DDU cost is 2 log records x 0.0005 log weight = 0.001 DDUs.

FAQs

Can I change the retention period for log monitoring?

Beginning with Dynatrace version 1.208+, there is no longer a way to change retention periods in SaaS. This is however possible for Dynatrace Managed accounts.

Where can I check my log consumption for DDUs?

To check your DDU consumption

In the Dynatrace menu, go to Consumption. Note: You need the Change monitoring settings permission or an admin account to access this page.
On the Davis data units (DDU) page:
- In the Consumption by DDU pool table, check the Log Monitoring row.
- In the DDU consumption details section, check the Log Monitoring tab.

What is the weight of one log record?

Each log record consumes 0.0005 DDU.

What is the size of one log record?

Dynatrace recognizes a log record as a single log entry that contains a valid timestamp or is described in a JSON object. This log record can contain multiple lines of log data and, depending on your log verbosity, the same amount of log data may contain a different number of log records consuming DDUs.

Are there any host-included DDUs available for Log Monitoring?

No. Dynatrace only offers host-included metrics. Log ingestion always consumes DDUs, which are deducted from your available quota.