HTTP Event Collector (HEC)
LogScale's HEC API is an alternative HTTP ingest endpoint. You will find
it at /api/v1/ingest/hec
and at /services/collector
.
Note
The best practices described in Ingest via API Best Practices also apply to the HEC API.
Format of Data
Ingested data is a series of whitespace delimited JSON objects, containing one or more of the following elements. All elements are optional.
Member | Description |
---|---|
time
|
Time in seconds since January 1, 1970 in UTC. This can be a integer or floating point number to support milliseconds. LogScale represents time with millisecond precision. Times are interpreted as follows:
|
timezone
|
Can be used to describe the time zone in which the event
happened. Defaults to
Z (i.e., UTC).
|
index
| Optional name of the repository to ingest into. In public-facing API's this must — if present — be equal to the repository used to create the ingest token used for authentication. In private cluster setups, humio can be configured to allow these to be different. See below. |
sourcetype
| Translated to #type inside LogScale. If set, this is used to choose which LogScale parser to use for extracting fields. |
source
| Translated to the @source field in LogScale. Typically used to designate the path to the file that is being shipped to LogScale. |
host
| Translated to the @host field in LogScale. Typically used to designate the origin host. |
event
|
This can be either a JSON Object or a String. Translated to the
@rawstring field in LogScale. When this is
a JSON Object, all members of the object will become accessible
fields in humio with no further processing. If it is a string,
the key/value parser is always applied to the string to extract
elements. The key/value parser searches for
key=value ,
key= ” value “
or key='value' .
|
fields
|
JSON object containing extra fields to the event. This can be
used if event is a string,
boolean or number and it is pre-processed prior to ingest to
extract fields. Tags #tags can be added to
the event by specifying fields starting with
# . It cannot be used if
event is an array or an
object.
|
Authentication
You will need to provide a Ingest Tokens in
the HTTP Authorization
header.
The ingest token contains the name of the repository the data is stored in, and ingested events will be stored in the repository corresponding to the ingest token.
If the LogScale configuration variable
ALLOW_CHANGE_REPO_ON_EVENTS
is set to true, then HEC
allows ingest to any repository specified as
"index": "<repository-name>"
in the body of a message, as long as the ingest token is valid for any
existing repository on the humio server. If the named repo does not
exist then the event remains in the repo designated by the ingest token.
This is a potential security issue on a public API endpoint, so this option should only be used inside a trusted environment.
Example
Below is an example of the contents of a json file,
events.json
:
{
"time" : 1537537729.0,
"event" : "Fri, 21 Sep 2018 13:48:49 GMT - system started name=webserver",
"source" : "/var/log/application.log",
"sourcetype" : "applog",
"fields" : { "#env" : "prod" }
}
{
"time" : 1537535729.0,
"event" : {
"message" : "System shutdown",
"host" : { "ip" : "127.0.0.1", "port" : 2222 }
},
"fields" : { "#datacenter" : "amazon-east1" }
}
curl $YOUR_LOGSCALE_URL/api/v1/ingest/hec \
-X POST \
-H "Content-Type: text/plain; charset=utf-8" \
-H "Authorization: Bearer $INGEST_TOKEN" \
--data "@events.json"
Note
The Content-Type
supports both
text/plain
and
application/json
You must make the following changes to the sample configuration
Add other fields in the fields section. These fields, and their values, will be added to each event.
Insert the URL containing the LogScale host in the
$YOUR_LOGSCALE_URL
Endpoints field.$INGEST_TOKEN
is the Ingest Tokens for your repository, (a string such asfS6Kdlb0clqe0UwPcc4slvNFP3Qn1COzG9DEVLw7v0Ii
).
Raw HEC
LogScale's Raw HEC API is a simple line-delimited ingest endpoint for unstructured logs.
You will find it at /api/v1/ingest/hec/raw
and at
/services/collector/raw
.
Simply send a POST to one of the two endpoints above. Each line in the
input (separated by /n
,
/r
, or
/r/n
) will be ingested as an
event.
You can optionally add an
X-Splunk-Request-Channel
header or
channel
as a query parameter.
These will be added as a field on the event named "channel".
Example
In a simple text editor, create a file named,
events.txt
and copy the
following lines into it:
Fri, 21 Sep 2018 13:48:49 GMT - system started name=webserver
System shutdown
Then execute the following from the command-line:
curl $YOUR_LOGSCALE_URL/api/v1/ingest/hec/raw?channel=foo \
-X POST \
-H "Content-Type: text/plain; charset=utf-8" \
-H "Authorization: Bearer $INGEST_TOKEN" \
--data-binary "@events.txt"
You must make the following changes to the sample configuration:
Insert the URL containing the LogScale host in the $YOUR_LOGSCALE_URL field.
$INGEST_TOKEN
is the Ingest Tokens for your repository, (a string such asfS6Kdlb0clqe0UwPcc4slvNFP3Qn1COzG9DEVLw7v0Ii
).