IBM Cloud™ Internet Services on Enterprise Plan offers a Logpush feature which sends at least 1 log package (on a .gz
file) to a Bucket on a IBM Cloud Object Storage every 100,000 logs or every 30 seconds, whichever comes first. More than one file might be pushed per 30-second period or per 100,000 logs. For those logs, there is a service called IBM Log Analysis with LogDNA that can receive all logs and display them in a single platform (you can send logs from your Kubernetes cluster, VMs, etc). To import all logs into LogDNA, you need to set up a Serverless function on IBM Cloud Functions which uses a Trigger to call your function automatically. The Trigger listens for a write event on IBM Cloud Object Storage. Whenever CIS uploads an object to your IBM Cloud Object Storage bucket, the Trigger calls your function that process the log package and automatcally send it to your LogDNA instance.
Before you follow step-by-step below, you need to install IBM Cloud CLI and IBM Cloud Functions CLI in your local machine. Then, you need to login in your IBM Cloud account on IBM Cloud CLI (if you haven't already done, run ibmcloud login
).
Download the source code from Github and access the project folder.
git clone https://github.com/IBM/logdna-cos.git
cd logdna-cos
Access the IBM Cloud Catalog and create a IBM Cloud Object Storage. After you create the instance, you have to create two Buckets with the same Resiliency and Location (e.g Regional
and us-south
):
- To receive the log files from CIS;
- To store the log files after you send the content to LogDNA.
Remember the name for each one of the Bucket name, because you're going to use them in the next step.
For your service instance, you need to create a Service credential. You can find it on the left menu in your COS instance. For the purpose of this project, you will use apikey
and iam_serviceid_crn
.
You can find the Endpoint URL on
Endpoints
tab. The correct enpoint for your usecase depends on the Resilience and Location you choose when you create your Buckets. For more information, access the IBM Cloud Docs.
Access the IBM Cloud Catalog and create a IBM Log Analysis with LogDNA. After you create the instance, you have to access the service by clicking on View LogDNA
button.
Access the Settings
-> ORGANIZATION
-> API Keys
to get your Ingestion Keys.
Run the following command with the IBM Cloud Object Storage credentials and the bucket name (for long-term retention), and IBM Log Analysis with LogDNA ingestion key:
- LOGDNA_HOSTNAME is the name of the source of the log line.
- LOGDNA_INGESTION_KEY is used to connect the Node.js function to the LogDNA instance.
- COS_BUCKET_ARCHIVE is the bucket where you will save the log package after you send it to LogDNA (consider it as your long-term retention).
- COS_APIKEY is the apikey field, generated on service credentials in your COS instance.
- COS_ENDPOINT is the endpoint available on Endpoint section in your COS instance. It depends on the resiliency and location that your bucket is defined.
- COS_INSTANCEID is the resource_instance_id field, generated on service credentials in your COS instance.
export LOGDNA_HOSTNAME="" LOGDNA_INGESTION_KEY="" COS_BUCKET_ARCHIVE="" COS_APIKEY="" COS_ENDPOINT="" COS_INSTANCEID=""
Run the following command to deploy handler.js
function and to set up the Trigger with Cron job.
As you are using IBM Cloud Functions, you don't need to install any package or setup a
package.json
. The platform already has all libraries required to run the source code.
ibmcloud fn deploy --manifest manifest.yml
Before you can create a trigger to listen for bucket change events, you must assign the Notifications Manager role to your IBM Cloud Functions namespace. Folow the instruction on IBM Cloud Docs to assign the role in your service policy.
When you complete the previous step, you will be able to create a new COS Trigger. Access the Functions > Create > Trigger > Cloud Object Storage to create and connect your bucket with your Action.
Now, your Action will be called everytime you upload a new object to your bucket.
To set up the Logpush on Cloud Internet Services (CIS) and authorize CIS to send the log package to Object Storage, follow the step-by-step from IBM Cloud Docs.
Cloud Internet Services on Enterprise plan (Enterprise Package or Enterprise Usage) allows users to have access to detailed logs of HTTP and range requests for their domains. These logs help debug and analytics. A high volume of access by a given user can mean a possible DoS (Denial of Service) attack.
{
"CacheCacheStatus": "unknown",
"CacheResponseBytes": 0,
"CacheResponseStatus": 0,
"CacheTieredFill": false,
"ClientASN": 12876,
"ClientCountry": "fr",
"ClientDeviceType": "desktop",
"ClientIP": "212.47.254.129",
"ClientIPClass": "noRecord",
"ClientRequestBytes": 977,
"ClientRequestHost": "ibmcloud.xyz",
"ClientRequestMethod": "GET",
"ClientRequestPath": "/",
"ClientRequestProtocol": "HTTP/1.1",
"ClientRequestReferer": "http://ibmcloud.xyz/",
"ClientRequestURI": "/",
"ClientRequestUserAgent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.89 Safari/537.36",
"ClientSSLCipher": "NONE",
"ClientSSLProtocol": "none",
"ClientSrcPort": 41978,
"ClientXRequestedWith": "",
"EdgeColoCode": "CDG",
"EdgeColoID": 19,
"EdgeEndTimestamp": "2020-08-05T22:06:14Z",
"EdgePathingOp": "ban",
"EdgePathingSrc": "filterBasedFirewall",
"EdgePathingStatus": "nr",
"EdgeRateLimitAction": "",
"EdgeRateLimitID": 0,
"EdgeRequestHost": "",
"EdgeResponseBytes": 1864,
"EdgeResponseCompressionRatio": 2.65,
"EdgeResponseContentType": "text/html",
"EdgeResponseStatus": 403,
"EdgeServerIP": "",
"EdgeStartTimestamp": "2020-08-05T22:06:14Z",
"FirewallMatchesActions": ["drop"],
"FirewallMatchesRuleIDs": ["570fd13c23eb4b49bcd8d585df59d14f"],
"FirewallMatchesSources": ["firewallRules"],
"OriginIP": "",
"OriginResponseBytes": 0,
"OriginResponseHTTPExpires": "",
"OriginResponseHTTPLastModified": "",
"OriginResponseStatus": 0,
"OriginResponseTime": 0,
"OriginSSLProtocol": "unknown",
"ParentRayID": "00",
"RayID": "5be3d2fdfc3fcd97",
"SecurityLevel": "unk",
"WAFAction": "unknown",
"WAFFlags": "0",
"WAFMatchedVar": "",
"WAFProfile": "unknown",
"WAFRuleID": "",
"WAFRuleMessage": "",
"WorkerCPUTime": 0,
"WorkerStatus": "unknown",
"WorkerSubrequest": false,
"WorkerSubrequestCount": 0,
"ZoneID": 259368151
}
After you set up the Logpush feature, it will start sending the log package to your IBM Cloud Object Storage instance every 5 minutes, whether there was an access or not. For a large amount of access, you will see more than 1 log package (the .gz
file) on COS. That is because Cloudflare splits the file to avoid a large file size (more than 3 GB per log package).
After you set up all the steps above, you will be able to see the logs being imported on LogDNA, as you can see on the image below.
- LogDNA ingestion API has a limitation of 10 MB per request.
- ESOCKETTIMEDOUT, ECONNRESET and ETIMEDOUT are LogDNA Ingest API errors. The script will automatically resend the logs.
Copyright 2021 IBM Corp.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.