CoreDNS Log Dashboard User Guide

Last updated: 2024-11-04 10:20:24

Tencent Kubernetes Engine (TKE) has deployed CoreDNS to provide domain name resolution service within a cluster. Due to various reasons such as network failures or excessive CoreDNS load, DNS request exception, high request latency, and uneven distribution of CoreDNS requests among multiple replicas may occur, thereby affecting users' normal DNS requests. To quickly troubleshoot DNS exception and identify potential business and security vulnerabilities, TKE has built a comprehensive CoreDNS logging capability based on the CoreDNS log plugin and the Cloud Log Service (CLS) log platform. This document will guide you on how to enable CoreDNS logs in a TKE cluster and use the corresponding dashboard feature for troubleshooting.
Prerequisites
1. CLS should be activated for clusters.
2. The log plugin needs to be added to the Corefile configuration of CoreDNS.
Note:
Add the log plugin to the Corefile configuration as follows, and edit the configmap named coredns under kube-system.
 data:
  Corefile: |2-
        .:53 {
            template ANY HINFO . {
                rcode NXDOMAIN
            }
            log  # Add the log plugin here.
            errors
            health {
                lameduck 30s
            }
            ready
            kubernetes cluster.local. in-addr.arpa ip6.arpa {
                pods insecure
                fallthrough in-addr.arpa ip6.arpa
            }
            prometheus :9153
            forward . /etc/resolv.conf {
                prefer_udp
            }
            cache 30
            reload
            loadbalance
        }
kind: ConfigMap
Save the configuration and exit. The Corefile will be automatically reloaded. If the Corefile is not configured for reloading, you need to rebuild CoreDNS to make the configuration effective.
3. Ensure the cluster's CoreDNS version is 1.8.4 or later. If you need to upgrade CoreDNS to version 1.8.4, refer to Upgrading to v1.8.4.
Enabling CoreDNS Logs
1. Log in to the TKE console and select O&M Feature Management in the left sidebar.
2. Select the cluster for which you want to enable CoreDNS logs and click Settings on the right side of the cluster, as shown in the figure below:
﻿
3. On the Set feature page, click Edit to the right of Log Collection.
4. Select Enable Log Collection and click Confirm, as shown in the figure below:
Note:
If Step 2 in Prerequisites is not completed, the enabling operation cannot be performed.
﻿
5. Click Edit to the right of Network Logs, as shown in the figure below:
﻿
6. Select Enable CoreDNS Logs and enter the following information:
﻿
Log region: Select a region for storing CLS log sets.
Log set: Select a CLS log set name. If there is no suitable log set, you can create a log set.
Log topic: You can choose to automatically create a log topic or select an existing log topic.
7. Click Confirm to enable CoreDNS logs.
﻿
Click the log topic link to enter the CLS page to query logs and perform other operations. The meanings of the log index fields are as follows:
Field Name
Description
Example
class
Request category.
IN
do
Whether "DNSSEC OK" (Domain Name System Security Extensions Confirmation) is set in a query.
false
duration
Response time (in seconds).
0.000098921
id
Request ID, which identifies a specific DNS request and response.
30008
level
Log level.
INFO
name
Target domain name queried in a DNS request.
craned.crane-system.svc.cluster.local.
port
Client port sending a DNS request.
50424
proto
Protocol used.
udp
rcode
Response code.
NXDOMAIN
remote
Client IP address.
10.99.10.128
rflags
Flag fields in response messages, which indicate the status and results of a DNS query.
qr, aa, rd
rsize
Maximum DNS response size.
162
size
Maximum DNS request size.
69
bufsize
Internal buffer size for DNS requests and responses.
65535
type
Request type.
A
Using the CoreDNS Dashboard in Log Management
1. Log in to the TKE console and select Log Management > CoreDNS Logs in the left sidebar.
2. Go to the CoreDNS Log page and select the region, cluster type, and the cluster you need to view, as shown in the figure below:
﻿
3. View dashboard data, as shown in the figure below:
﻿
Request Success Rate: Calculates the proportion of all normal DNS responses (NOERROR and NXDOMAIN) to the total number of requests. You can use this metric to identify whether there are any resolution failures in the current CoreDNS.
Number of Domains: Displays the total number of domain names responded to by the current CoreDNS service.
Request QPS: Reflects the queries per second (QPS) performance of CoreDNS service over a certain time period . You can use the sequence diagram to identify performance issues in CoreDNS.
Average Latency/P95 Latency/P99 Latency: Reflects the average latency, P95 latency, and P99 latency of the last 10,000 requests in the CoreDNS service, helping to identify slow response issues in CoreDNS.
CoreDNS Pod Request Distribution: Displays the request distribution and average latency for each replica in multi-replica CoreDNS scenarios, helping to identify issues with uneven request distribution among CoreDNS replicas.
Slow Resolution Log: Records relevant information in the slow resolution log when DNS request processing time exceeds a specific threshold. By analyzing the slow resolution log, you can identify the types of requests that take the most time and optimize accordingly.
Disabling CoreDNS Logs
If you no longer need CoreDNS log collection, you can disable CoreDNS log collection capability as follows:
1. Log in to the TKE console and select O&M Feature Management in the left sidebar.
2. Select the cluster for which you need to disable CoreDNS logs and click Settings on the right side of the cluster.
3. On the Set feature page, click Edit to the right of Network Log, as shown in the figure below:
﻿
4. Deselect Enable CoreDNS Logs, as shown in the figure below:
﻿
5. Click Confirm. If a log topic is automatically created, you will be prompted about the associated log topic. If you no longer need this log topic, click to go to the CLS console to delete the corresponding log topic. Otherwise, the associated log topic will be retained and incur charges.
﻿
﻿

tencent cloud

Recent Pages

CoreDNS Log Dashboard User Guide

Prerequisites

Enabling CoreDNS Logs

Using the CoreDNS Dashboard in Log Management

Disabling CoreDNS Logs

Was this page helpful?

Was this page helpful?

Field Name	Description	Example
class	Request category.	IN
do	Whether "DNSSEC OK" (Domain Name System Security Extensions Confirmation) is set in a query.	false
duration	Response time (in seconds).	0.000098921
id	Request ID, which identifies a specific DNS request and response.	30008
level	Log level.	INFO
name	Target domain name queried in a DNS request.	craned.crane-system.svc.cluster.local.
port	Client port sending a DNS request.	50424
proto	Protocol used.	udp
rcode	Response code.	NXDOMAIN
remote	Client IP address.	10.99.10.128
rflags	Flag fields in response messages, which indicate the status and results of a DNS query.	qr, aa, rd
rsize	Maximum DNS response size.	162
size	Maximum DNS request size.	69
bufsize	Internal buffer size for DNS requests and responses.	65535
type	Request type.	A

tencent cloud

Sign Up

Log in

Recent Pages

CoreDNS Log Dashboard User Guide

Prerequisites

Enabling CoreDNS Logs

Using the CoreDNS Dashboard in Log Management

Disabling CoreDNS Logs

Was this page helpful?

Was this page helpful?