Tagged: Docker Toggle Comment Threads | Keyboard Shortcuts

  • Wang 21:20 on 2019-04-19 Permalink | Reply
    Tags: Docker,   

    Compile consul source code and generate executable scripts 

    Recently I have some special requirements which ask to compile consul source code and generate executable scripts, below are the general steps:

    1. Make sure you have installed docker, go, nodejs

    xxx@xxx consul $ docker version && node --version && go version
    Client: Docker Engine - Community
     Version:           18.09.2
     API version:       1.39
     Go version:        go1.10.8
     Git commit:        6247962
     Built:             Sun Feb 10 04:12:39 2019
     OS/Arch:           darwin/amd64
     Experimental:      false
    Server: Docker Engine - Community
      Version:          18.09.2
      API version:      1.39 (minimum version 1.12)
      Go version:       go1.10.6
      Git commit:       6247962
      Built:            Sun Feb 10 04:13:06 2019
      OS/Arch:          linux/amd64
      Experimental:     false
    go version go1.12.4 darwin/amd64

    2. Download consul source code

    cd $GO_HOME && mkdir -p src/github.com/hashicorp
    cd src/github.com/hashicorp
    git clone https://github.com/hashicorp/consul.git
    cd consul && git checkout v1.4.4 -b <BRANCH_NAME>

    3. Modify UI templates under ui-v2/app/templates/

    4. Compile & generate executable scripts

    export ALLOW_DIRTY_GIT=1
    sh build-support/scripts/release.sh --tag false --sign false --release <RELEASE_NAME>

    5. Test consul

    bin/consul agent 


    1. generate consul packages locations: bin/consul, pkg/bin, pkg/dist
    2. reference: https://github.com/hashicorp/consul/blob/master/.github/CONTRIBUTING.md
  • Wang 22:08 on 2019-03-07 Permalink | Reply
    Tags: Docker, , ,   

    Consul with ACL 

    Enable ACL in Consul to protect your configurations, I deployed Consul by Helm.

  • Wang 22:12 on 2019-02-11 Permalink | Reply
    Tags: , , , Docker, , , , , ,   

    Guarantee service availability in kubernetes 

    A good service not only provide good functionalities, but also ensure the availability and uptime.

    We reinforce our service from QoS, QPS, Throttling, Scaling, Throughput, Monitoring.


    There’re 3 kinds of QoS in kubernetes: Guaranteed, Burstable, BestEffort. We usually use Guaranteed, Burstable for different services.

        cpu: 1000m
        memory: 4Gi
        cpu: 1000m
        memory: 4Gi
        cpu: 1000m
        memory: 4Gi
        cpu: 6000m
        memory: 8Gi

    We did lots of stress test on APIs by Gatling before we release them, we mainly care about mean response time, std deviation, mean requests/sec, error rate (API Testing Report), during testing we monitor server metrics by Datadog to find out bottlenecks.

    We usually test APIs in two scenarios: internal, external. External testing result is much lower than internal testing because of network latency, network bandwidth and son on.

    Internal testing result

    ---- Global Information --------------------------------------------------------
    > request count                                     246000 (OK=246000 KO=0     )
    > min response time                                     16 (OK=16     KO=-     )
    > max response time                                   5891 (OK=5891   KO=-     )
    > mean response time                                    86 (OK=86     KO=-     )
    > std deviation                                        345 (OK=345    KO=-     )
    > response time 50th percentile                         30 (OK=30     KO=-     )
    > response time 75th percentile                         40 (OK=40     KO=-     )
    > response time 95th percentile                         88 (OK=88     KO=-     )
    > response time 99th percentile                       1940 (OK=1940   KO=-     )
    > mean requests/sec                                817.276 (OK=817.276 KO=-     )
    ---- Response Time Distraaibution ------------------------------------------------
    > t < 800 ms                                        240565 ( 98%)
    > 800 ms < t < 1200 ms                                1110 (  0%)
    > t > 1200 ms                                         4325 (  2%)
    > failed                                                 0 (  0%)

    External testing result

    ---- Global Information --------------------------------------------------------
    > request count                                      33000 (OK=32999  KO=1     )
    > min response time                                    477 (OK=477    KO=60001 )
    > max response time                                  60001 (OK=41751  KO=60001 )
    > mean response time                                   600 (OK=599    KO=60001 )
    > std deviation                                        584 (OK=484    KO=0     )
    > response time 50th percentile                        497 (OK=497    KO=60001 )
    > response time 75th percentile                        506 (OK=506    KO=60001 )
    > response time 95th percentile                       1366 (OK=1366   KO=60001 )
    > response time 99th percentile                       2125 (OK=2122   KO=60001 )
    > mean requests/sec                                109.635 (OK=109.631 KO=0.003 )
    ---- Response Time Distribution ------------------------------------------------
    > t < 800 ms                                         29826 ( 90%)
    > 800 ms < t < 1200 ms                                1166 (  4%)
    > t > 1200 ms                                         2007 (  6%)
    > failed                                                 1 (  0%)
    ---- Errors --------------------------------------------------------------------
    > i.g.h.c.i.RequestTimeoutException: Request timeout after 60000      1 (100.0%)

    We throttle API by Nginx limit, we configured ingress like this:

      nginx.ingress.kubernetes.io/limit-connections: '30'
      nginx.ingress.kubernetes.io/limit-rps: '60'

    And it will generate Nginx configuration dynamically like this:

    limit_conn_zone $limit_ZGVsaXZlcnktY2RuYV9kc2QtYXBpLWNkbmEtZ2F0ZXdheQ zone=xxx_conn:5m;
    limit_req_zone $limit_ZGVsaXZlcnktY2RuYV9kc2QtYXBpLWNkbmEtZ2F0ZXdheQ zone=xxx_rps:5m rate=60r/s;
    server {
        server_name xxx.xxx ;
        listen 80;
        location ~* "^/xxx/?(?<baseuri>.*)" {
            limit_conn xxx_conn 30;
            limit_req zone=xxx_rps burst=300 nodelay;

    We use HPA in kubernetes to ensure auto (Auto scaling in kubernetes), you could check HPA status in server:

    [xxx@xxx ~]$ kubectl get hpa -n test-ns
    api-demo   Deployment/api-demo   39%/30%, 0%/30%   3         10        3          126d
    [xxx@xxx ~]$ kubectl get pod -n test-ns
    NAME                           READY     STATUS    RESTARTS   AGE
    api-demo-76b9954f57-6hvzx      1/1       Running   0          126d
    api-demo-76b9954f57-mllsx      1/1       Running   0          126d
    api-demo-76b9954f57-s22k8      1/1       Running   0          126d
    Throughput & Monitoring

    We integrated Datadog for monitoring(Monitoring by Datadog), we could check detail API metrics from various dashboards.

    Also we could calculate throughout from user, request, request time.

  • Wang 21:26 on 2019-01-14 Permalink | Reply
    Tags: , , , Docker, , ,   

    Monitoring by Datadog 

    We have thousands of containers running on hundreds of servers, so we need comprehensive monitoring system to monitor service and server metrics.

    We investigated popular cloud monitoring platform: New Relic and Datadog, finally we decided to use datadog.

    Dashboard: Datadog could  detect services and configure dashboards for you automatically.

    Container & Process: You could check all your containers & process in all environments clearly.

    Monitors: Datadog will create monitors according to service type automatically, if it doesn’t your requirement, you could create your own. It’s also convenient to send alert message through Slack, Email.

    APM: Datadog provide various charts for API analysis, also there’s Service Map which you could check service dependencies.

    Synthetics: New feature in Datadog which could test your API around the world to check availability and uptime.

  • Wang 21:44 on 2018-11-20 Permalink | Reply
    Tags: , , , Docker, , ,   

    Sticky session in Kubernetes 

    As we know RESTful API is stateless, every request will be forward to backend server by round robin mechanism.

    But in some scenario we need sticky session which means request from one client should be forward to one backend server.

    After checking kubernetes documentation we added some annotations under ingress configuration, and it works well.

      nginx.ingress.kubernetes.io/affinity: "cookie"
      nginx.ingress.kubernetes.io/session-cookie-name: "router"
      nginx.ingress.kubernetes.io/session-cookie-hash: "sha1"

    If you open Developer Tools in Chrome, you will find the cookie.

  • Wang 22:21 on 2018-11-05 Permalink | Reply
    Tags: , , , Docker, , , , ,   

    [Presto] Secure with LDAP 

    For security issue we decided to enable LDAP in presto, to deploy presto into kubernetes cluster we build presto image ourselves which include kerberos authentication and LDAP configurations.

    As you see the image structure, configurations under catalog/etc/hive are very important, please pay attention.

    krb5.conf and xxx.keytab are used to connect to kerberos

    password-authenticator.properties and ldap_server.pem under etc, hive.properties and hive-security.json under catalog are used to connect to LDAP.




    hive.config.resources=core-site.xml, hdfs-site.xml


      "schemas": [{
        "user": "user_1",
        "schema": "db_1",
        "owner": false
      }, {
        "user": " ",
        "schema": "db_1",
        "owner": false
      }, {
        "user": "user_2",
        "schema": "db_2",
        "owner": false
      "tables": [{
        "user": "user_1",
        "schema": "db_1",
        "table": "table_1",
        "privileges": ["SELECT"]
      }, {
        "user": "user_1",
        "schema": "db_1",
        "table": "table_2",
        "privileges": ["SELECT"]
      }, {
        "user": "user_2",
        "schema": "db_1",
        "table": ".*",
        "privileges": ["SELECT"]
      }, {
        "user": "user_2",
        "schema": "db_2",
        "table": "table_1",
        "privileges": ["SELECT"]
      }, {
        "user": "user_2",
        "schema": "db_2",
        "table": "table_2",
        "privileges": ["SELECT"]
      "sessionProperties": [{
        "allow": false
  • Wang 22:30 on 2018-10-15 Permalink | Reply
    Tags: , , Docker, ,   

    Jenkins pipeline & kubernetes 

    We build deployment pipeline by Jenkins, Git, Maven, Docker, JFrog, Kubernetes, Slack, below is overall process:

    develop -> create branch -> push code -> git hook -> jenkins build -> code check -> unit test -> docker build -> push docker image -> deploy -> notificationa

    For every project we generate pipeline scripts by JHipster like this:

    ci contains docker related scripts, cd contains kubernetes related scripts.

    We configured Jenkins to scan projects from git automatically which followed naming rule, if any changes on git, Jenkins will pull the code and start building.

  • Wang 22:43 on 2018-10-08 Permalink | Reply
    Tags: , , , Docker, , , ,   

    Nginx ingress in kubernetes 

    There are 3 ways to expose your service: NodePort, LoadBalancer, Ingress, next I will introduce about how to use ingress.

    1.Deploy ingress controller

    You need deploy ingress controller at first which will start nginx pods, then nginx will bind domains and listen to the requests.

    I built a common ingress chart for different service, I only need change values-<service>.yaml and deploy script if any changes.

    Another key point is that you must be clear about ingress-class, different service use different ingress-class, it will be quite messy if you mistake them.

      - /nginx-ingress-controller
      - --default-backend-service=$(POD_NAMESPACE)/default-http-backend
      - --configmap=$(POD_NAMESPACE)/nginx-configuration
      - --tcp-services-configmap=$(POD_NAMESPACE)/tcp-services
      - --udp-services-configmap=$(POD_NAMESPACE)/udp-services
      - --ingress-class={{ .Values.server.namespace }}
      - --sort-backends=true

    2.Configure service ingress

    Next we need configure service ingress which will append nginx server configuration dynamically.

    I also built a service chart which include environment configurations, Jenkins & Helm will use different values-<env>.yaml when execute pipeline deployment.

    Ingress example:

    apiVersion: extensions/v1beta1
    kind: Ingress
      name: {{ .Values.app.name }}{{ .Values.deploy.subfix }}
      namespace: {{ .Values.app.namespace }}
        kubernetes.io/ingress.class: "{{ .Values.ingress.class }}"
        kubernetes.io/tls-acme: "true"
        nginx.ingress.kubernetes.io/enable-cors: "false"
        nginx.ingress.kubernetes.io/rewrite-target: /
        nginx.ingress.kubernetes.io/proxy-body-size: 10m
      - host: {{ .Values.ingress.hostname }}
          - path: {{ .Values.ingress.path }}
              serviceName: {{ .Values.app.name }}{{ .Values.deploy.subfix }}
              servicePort: {{ .Values.container.port }}
  • Wang 21:23 on 2018-09-06 Permalink | Reply
    Tags: , , Docker, ,   

    Probe in kubernetes 

    There’s two kinds of probe: readinessProbe, livenessProbe in kubernetes used to detect if your service is healthy.

    We encountered a problem when configured readinessProbe, there’s a property named initialDelaySeconds which indicate kubernetes will start health check after specific second, we used the default value 60 which means kubernetes will check health after 60 seconds.

      initialDelaySeconds: 60
      timeoutSeconds: 5

    As we deployed over 20 StatefulSet pods and these pods joined as a cluster which cost over 60 seconds, kubernetes can’t ping service successfully so that kubernetes restart these pods, thees pods restart in loop all the time.

    After we increased the initialDelaySeconds to 120, everything goes fine.

  • Wang 21:56 on 2018-08-22 Permalink | Reply
    Tags: , , Docker, ,   

    Stateful deployment in kubernetes 

    If you deploy pod by setting “kind: Deployment“, you will lost your data when the pod restart or being deleted.

    It’s not acceptable when we want to deploy storage system like Redis, Elasticsearch, in this case we need use StatefulSet.

    For the concrete explanation please refer to the official documentation, StatefulSet use PVC(Persistent Volume Claim) as storage, and it will exist all the time no matter what happened to the pod.

    You must specify PVC in StatefulSet’s yaml file like this:

    - metadata:
      name: redis
      accessModes: [ "ReadWriteOnce" ]
      storageClassName: fast
          storage: 10Gi

    Please also pay attention to PVC’s name, there’s a rule for StatefulSet and PVC name mapping which IS NOT covered by documentation.

Compose new post
Next post/Next comment
Previous post/Previous comment
Show/Hide comments
Go to top
Go to login
Show/Hide help
shift + esc
%d bloggers like this: