I am using spring config server with 2 backends : git and vault (for secrets), and i have a clients apps that connect to the config server to get distant configuration (git and vault).
I have this configuration:
config server
server:
port: 8888
spring:
profiles:
active: git, vault
cloud:
config:
server:
vault:
host: hostName
kvVersion: 1
order: 1
backend: secret/cad
scheme: https
port: 443
git:
order: 2
uri: git#gitlab.git_repo
ignoreLocalSshSettings: true
force-pull: true
deleteUntrackedBranches: true
privateKey: key
and client side
spring:
application:
name: my_app_name
cloud:
vault:
config:
uri: http://localhost:8888
token: s.token
fail-fast: true
With this way I have to change the token for every client every day (token expire 24h). Is there a way to renew the token with this configuration or there is another way to authenticate to the vault?
spring.cloud.vault:
config.lifecycle:
enabled: true
min-renewal: 10s
expiry-threshold: 1440m
lease-endpoints: Legacy
1440 minutes = 24h
Reference: https://cloud.spring.io/spring-cloud-vault/reference/html/#vault-lease-renewal
Related
In my fresh Jhipster Registry 7.0.4 launch with jar application from github (with dev profile), I can see the logs menu (and all other).
I apply the default configuration files.
Application-dev.yml:
spring:
devtools:
restart:
enabled: false
livereload:
enabled: false
mail:
host: localhost
port: 25
username:
password:
security:
user:
name: admin
password: admin
role: USER,ADMIN
server:
port: 9762
compression:
enabled: true
mime-types: text/html,text/xml,text/plain,text/css, application/javascript, application/json
min-response-size: 1024
eureka:
client:
service-url:
defaultZone: http://admin:admin#localhost:9762/eureka
logging:
file: logs/jhipster-registry.log
level:
ROOT: INFO
io.github.jhipster: INFO
io.github.jhipster.registry: INFO
jhipster:
http:
cache: # Used by the CachingHttpHeadersFilter
timeToLiveInDays: 1461
security:
authentication:
jwt:
token-validity-in-seconds: 86400
token-validity-in-seconds-for-remember-me: 2592000
mail: # specific JHipster mail property, for standard properties see MailProperties
from: JHipsterRegistry#localhost
base-url: http://my-server-url-to-change # Modify according to your server's URL
logging:
logstash: # Forward logs to logstash over a socket, used by LoggingConfiguration
enabled: false
host: localhost
port: 5000
queue-size: 512
cache:
hazelcast:
# application:
registry is OK I can use it but I can't see and configure log level of jhipster of any module registered. How can I see them ?
I user SBA as discovery client (spring-boot-admin.version 2.1.5) to Eureka server. I see a registered application UP in Eureka server, but the same is DOWN on SBA.SBA app
These is my application.yml
spring:
application:
name: spring-boot-admin-sample-eureka
profiles:
active:
- secure
server:
port : 8091
tag::configuration-eureka[]
eureka: #<1>
instance:
leaseRenewalIntervalInSeconds: 10
statusPageUrlPath: /actuator/info
health-check-url-path: /actuator/health
metadata-map:
startup: ${random.int} #needed to trigger info and endpoint update after restart
management.context-path: /monitor-client/actuator
client:
registryFetchIntervalSeconds: 5
registerWithEureka: false
serviceUrl:
defaultZone: http://localhost:8761/ServicesRepository/eureka
management:
endpoints:
web:
exposure:
include: "*" #<2>
endpoint:
health:
show-details: ALWAYS
security:
enabled: false
end::configuration-eureka[]
spring:
profiles: insecure
spring:
profiles: secure
security:
user:
name: "user"
password: "password"
eureka:
instance:
metadata-map:
user.name: "user" #These two are needed so that the server
user.password: "password" #can access the protected client endpoints
spring:
boot:
admin:
client:
username: user
password: password
instance:
metadata:
user.name: user
user.password: password
Always show details in Health check section
management:
endpoint:
health:
show-details: "ALWAYS"
endpoints:
web:
exposure:
include: "*"
When i launch my server with prod profile, I can connect with admin admin in my JHipster application, but I can't save any data when I want to create a new object (I have an InternalServerError). However in dev profile with the same database (PostgreSQL) it's working well.
Moreover I don't know how to get logs when using prod profile to understand the problem.
I can add the configuration file for the prod profile
# ===================================================================
# Spring Boot configuration for the "prod" profile.
#
# This configuration overrides the application.yml file.
# ===================================================================
# ===================================================================
# Standard Spring Boot properties.
# Full reference is available at:
# http://docs.spring.io/spring-boot/docs/current/reference/html/common-application-properties.html
# ===================================================================
spring:
devtools:
restart:
enabled: false
livereload:
enabled: false
datasource:
url: jdbc:postgresql://localhost:5432/vraiJhipster
name:
username: vraiJhipster
password:
jpa:
database-platform: com.mycompany.myapp.domain.util.FixedPostgreSQL82Dialect
database: POSTGRESQL
show_sql: false
properties:
hibernate.cache.use_second_level_cache: true
hibernate.cache.use_query_cache: false
hibernate.generate_statistics: false
hibernate.cache.region.factory_class: org.hibernate.cache.ehcache.SingletonEhCacheRegionFactory
data:
elasticsearch:
cluster-name:
cluster-nodes: localhost:9300
mail:
host: localhost
port: 25
username:
password:
thymeleaf:
cache: true
liquibase:
contexts: prod
server:
port: 8080
compression:
enabled: true
mime-types: text/html,text/xml,text/plain,text/css, application/javascript, application/json
min-response-size: 1024
# ===================================================================
# JHipster specific properties
# ===================================================================
jhipster:
http:
cache: # Used by the CachingHttpHeadersFilter
timeToLiveInDays: 1461
cache: # Hibernate 2nd level cache, used by CacheConfiguration
timeToLiveSeconds: 3600
ehcache:
maxBytesLocalHeap: 256M
security:
rememberMe:
# security key (this key should be unique for your application, and kept secret)
key: *****************************************
mail: # specific JHipster mail property, for standard properties see MailProperties
from: vraiJhipster#localhost
metrics: # DropWizard Metrics configuration, used by MetricsConfiguration
jmx.enabled: true
spark:
enabled: false
host: localhost
port: 9999
graphite:
enabled: false
host: localhost
port: 2003
prefix: vraiJhipster
logs: # Reports Dropwizard metrics in the logs
enabled: false
reportFrequency: 60 # in seconds
logging:
logstash: # Forward logs to logstash over a socket, used by LoggingConfiguration
enabled: false
host: localhost
port: 5000
queueSize: 512
swagger: # swagger is disabled. It can be disabled by pasing 'no-swagger' profile at run time as well
enabled: false
Thank you.
If you need something about my configuration you can ask it.
I found the problem it was actually elasticSearch. When I use JHipster in production mode, the default configuration use elasticSearch on port 9300. But I hadn't any elasticSearch running on my computer. So I copy pasted the elasticSearch configuration from dev profile to prod profile.
data:
elasticsearch:
cluster-name:
cluster-nodes:
properties:
path:
logs: target/elasticsearch/log
data: target/elasticsearch/data
I am now trying to set up using spring config server / spring config client and eureka server to do service discovery together. when I was using local file system to save configuration, everything is fine, but as soon as I change to gitHub to save configuration, I can not start spring config client, and I got this error message:
com.netflix.discovery.DiscoveryClient : Can't get a response from http://user:{cipher}AQBdTe9vjgOqANyfi+Ke5HYmhFJnuvsGJ+8FOqLgdMdFInJkVRzy9XngylNp9KlX2lwPI13jT7ceDjBs9w4B6A15akQTHHP1q9rWwMcLtVCCKGFgvXSgQF+WWIItxBmIDGG4bSgDYZ6XQWl66T84PxjTBygErpmAjn+X2BWdtWtVmjXzpMQMaHGpiyLPoGIQYF4j9ZGpl16Jx79+j6et6LStksrT3WxQuanhZwIkqZ827+2mlUB3kd5ocW7R0vC3JMGVZlGNfqYQwJIhSJ19lrq4gxm/zbGroYonW3QSHTiRlhlvOdovEua/0dCl6+s0iEHYZCBllr7jikEWmS5Px7U2TG8/eQkjR8pUbSN6OE+OI2ZZUCNOT+mi75ZeeN1OtdQ=#localhost:8761/eureka/null
Can't contact any eureka nodes - possibly a security group issue?
java.lang.IllegalArgumentException: Illegal character in authority at index 7: http://user:{cipher}AQBdTe9vjgOqANyfi+Ke5HYmhFJnuvsGJ+8FOqLgdMdFInJkVRzy9XngylNp9KlX2lwPI13jT7ceDjBs9w4B6A15akQTHHP1q9rWwMcLtVCCKGFgvXSgQF+WWIItxBmIDGG4bSgDYZ6XQWl66T84PxjTBygErpmAjn+X2BWdtWtVmjXzpMQMaHGpiyLPoGIQYF4j9ZGpl16Jx79+j6et6LStksrT3WxQuanhZwIkqZ827+2mlUB3kd5ocW7R0vC3JMGVZlGNfqYQwJIhSJ19lrq4gxm/zbGroYonW3QSHTiRlhlvOdovEua/0dCl6+s0iEHYZCBllr7jikEWmS5Px7U2TG8/eQkjR8pUbSN6OE+OI2ZZUCNOT+mi75ZeeN1OtdQ=#localhost:8761/eureka/
at java.net.URI.create(URI.java:852)
This is what I have:
In cofig server application.yml:
spring:
application:
name: spirent-config-server
cloud:
config:
server:
git:
uri: https://github.com/spring-cloud-samples/config-repo
#searchPaths: foo,bar*
server:
port: 8888
in config client:
bootstrap.properties:
spring.application.name= foo
spring.profiles.active= development
spring.cloud.config.uri= http://localhost:8888
and application.yml
eureka:
instance:
hostname: ${host.instance.name:frank1}
nonSecurePort: ${host.instance.port:8080}
metadataMap:
instanceId: ${host.instance.instanceid:frank_123}
client:
serviceUrl:
defaultZone: http://localhost:8761/eureka/
server:
port: ${host.instance.port:8080}
Eureka Server:
application.yml
server:
port: 8761
eureka:
client:
registerWithEureka: false
fetchRegistry: false
server:
waitTimeInMsWhenSyncEmpty: 0
bootstrap.yml:
spring:
application:
name: eureka
cloud:
config:
uri: ${CONFIG_SERVER_URL:http://localhost:8888}
From the exception, it seemed it is because discoverClient is not working, anything special for the configuration?
But if I change from gitHub to local file storage for configuration, everything is fine.
Thanks for your suggestion in advance.
Based on the spring cloud document, I am doing like this:
In application.yml of config server:
spring:
application:
name: spirent-config-server
cloud:
config:
server:
git:
uri: https://github.com/spring-cloud-samples/config-repo
#searchPaths: foo,bar*
encrypt:
key: abc
server:
port: 8888
In application.yml of Eureka Server:
server:
port: 8761
eureka:
instance:
hostname: localhost
client:
registerWithEureka: false
fetchRegistry: false
serviceUrl:
defaultZone: http://${eureka.instance.hostname}:${server.port}/eureka/
server:
waitTimeInMsWhenSyncEmpty: 0
encrypt:
key: abc
And in application.yml of config client:
eureka:
instance:
hostname: ${host.instance.name:frank1}
nonSecurePort: ${host.instance.port:8080}
leaseRenewalIntervalInSeconds: 5 #default is 30, recommended to keep default
metadataMap:
#instanceId: ${host.instance.instanceid:frank_123}
instanceId: ${spring.application.name}:${spring.application.instance_id:${random.value}}
client:
serviceUrl:
defaultZone: http://localhost:8761/eureka/
server:
port: ${host.instance.port:8080}
encrypt:
key: abc
But, the problem is after I started Eureka an config server, it is fine, no exception, but after I started config client, I got this same exception:
Can't contact any eureka nodes - possibly a security group issue?
java.lang.IllegalArgumentException: Illegal character in authority at index 7: http://user:{cipher}AQBdTe9vjgOqANyfi+Ke5HYmhFJnuvsGJ+8FOqLgdMdFInJkVRzy9XngylNp9KlX2lwPI13jT7ceDjBs9w4B6A15akQTHHP1q9rWwMcLtVCCKGFgvXSgQF+WWIItxBmIDGG4bSgDYZ6XQWl66T84PxjTBygErpmAjn+X2BWdtWtVmjXzpMQMaHGpiyLPoGIQYF4j9ZGpl16Jx79+j6et6LStksrT3WxQuanhZwIkqZ827+2mlUB3kd5ocW7R0vC3JMGVZlGNfqYQwJIhSJ19lrq4gxm/zbGroYonW3QSHTiRlhlvOdovEua/0dCl6+s0iEHYZCBllr7jikEWmS5Px7U2TG8/eQkjR8pUbSN6OE+OI2ZZUCNOT+mi75ZeeN1OtdQ=#localhost:8761/eureka/
at java.net.URI.create(URI.java:852)
at com.sun.jersey.api.client.Client.resource(Client.java:433)
at com.netflix.discovery.DiscoveryClient.makeRemoteCall(DiscoveryClient.java:1091)
at com.netflix.discovery.DiscoveryClient.makeRemoteCall(DiscoveryClient.java:1060)
at com.netflix.discovery.DiscoveryClient.getAndStoreFullRegistry(DiscoveryClient.java:835)
Can anyone tell me where is my error? This only happened in using GIT as config server data source, but in local file data source, no error happened.
Thanks
These are my steps to try to resolve the key issues, but failed again:
download the "Java Cryptography Extension (JCE) Unlimited Strength Jurisdiction Policy Files" from Oracle, and follow instructions for installation (essentially replace the 2 policy files in the JRE lib/security directory with the ones that you downloaded).
Now if I run spring encrypt mysecret --key foo, it is successful, this means the JCE is installed correctly;
Creating a Key Store for Testing: using this:
https://github.com/spring-cloud/spring-cloud-config/blob/master/docs/src/main/asciidoc/spring-cloud-config.adoc#creating-a-key-store-for-testing
after updating the application.yml in my config server, the application.yml is this:
spring:
application:
name: spirent-config-server
cloud:
config:
server:
git:
uri: https://github.com/spring-cloud-samples/config-repo
#searchPaths: foo,bar*
encrypt:
keyStore:
location: classpath:/server.jks
password: letmein
alias: abc
secret: changeme
key: abc
server:
port: 8888
adding this to both eureka server and config client application.yml:
encrypt:
key: abc
Run everything:
Result is same nasty error
So what is wrong is my testing? Thanks for your help
It is my fault:
Changing application.yml to this:
spring:
application:
name: configserver
encrypt:
failOnError: false
keyStore:
#location: classpath:keystore.jks
#password: ${KEYSTORE_PASSWORD:foobar} # don't use a default in production
#alias: test
location: classpath:server.jks
password: letmein
alias: mytestkey
secret: changeme
And then using this command to generate the file resolving the issue completely:
keytool -genkeypair -alias mytestkey -keyalg RSA \
-dname "CN=Web Server,OU=Unit,O=Organization,L=City,S=State,C=US" \
-keypass changeme -keystore server.jks -storepass letmein
Thanks
I'm prototyping a set of Spring Cloud + Netflix OSS applications and have run into trouble with Eureka. In our setup, we have a Spring Cloud Config Server + Eureka Server, and then 2 modules that utilize that server component for bootstrapping and service discovery.
The problem I run into is that if I spin up 2 instances of the Eureka Server and try to pair them (based on the Two Peer Aware Eureka Servers in the docs) they don't synchronize with each other. See configs below and/or the code on GitHub.
Essentially, Peer1 starts up and looks fine. Peer2 will startup and look fine, with both peers showing each other in the services. However, if the "UserService" module spins up and registers itself with Peer1, Peer2 will never see it. If we then spin up the "Web" module pointing to Peer2, it can never resolve the UserService. They basically act in isolation.
I've tried several combinations of setting the serviceUrl both on the server and the instance of the Eureka servers but to no avail. Am I just configuring things wrong?
Peer 1 / default config:
server:
port: 8888
eureka:
dashboard:
path: /dashboard
instance:
hostname: peer1
leaseRenewalIntervalInSeconds: 3
client:
serviceUrl:
defaultZone: ${eureka.server.serviceUrl:http://localhost:${server.port}/eureka/}
server:
serviceUrl:
defaultZone: http://localhost:${server.port}/eureka/
peer2: http://peer2/eureka/
waitTimeInMsWhenSyncEmpty: 0
spring:
application:
name: demo-config-service
profiles:
active: native
# required for Spring Cloud Bus
rabbitmq:
host: ${DOCKER_IP:192.168.59.103}
port: 5672
username: guest
password: guest
virtualHost: /
cloud:
config:
server:
prefix: /configs
native:
searchLocations: /Users/dave/workspace/oss/distributed-spring/modules/config-server/src/main/resources/testConfigs
# git :
# uri: https://github.com/joshlong/microservices-lab-configuration
Peer 2 config:
server:
port: 8889
eureka:
dashboard:
path: /dashboard
instance:
hostname: peer2
leaseRenewalIntervalInSeconds: 3
client:
serviceUrl:
defaultZone: ${eureka.server.serviceUrl:http://localhost:${server.port}/eureka/}
server:
serviceUrl:
defaultZone: http://localhost:8888/eureka/
peer1: http://peer1/eureka/
waitTimeInMsWhenSyncEmpty: 0
spring:
application:
name: demo-config-service
profiles:
active: native
# required for Spring Cloud Bus
rabbitmq:
host: ${DOCKER_IP:192.168.59.103}
port: 5672
username: guest
password: guest
virtualHost: /
cloud:
config:
server:
prefix: /configs
native:
searchLocations: /Users/dave/workspace/oss/distributed-spring/modules/config-server/src/main/resources/testConfigs
# git :
# uri: https://github.com/joshlong/microservices-lab-configuration
There were a few problems. The defaultZone needs to be in the client section as noted in the docs. The defaultZone url needs the port.
/etc/hosts
127.0.0.1 peer1
127.0.0.1 peer2
Peer 1 Config (Partial)
eureka:
instance:
hostname: peer1
leaseRenewalIntervalInSeconds: 3
client:
serviceUrl:
defaultZone: http://peer2:8889/eureka/
Peer 2 Config (Partial)
eureka:
dashboard:
path: /dashboard
instance:
hostname: peer2
leaseRenewalIntervalInSeconds: 3
client:
serviceUrl:
defaultZone: http://peer1:8888/eureka/
server:
waitTimeInMsWhenSyncEmpty: 0
User service config (Partial) Config port was wrong.
spring:
application:
name: user-service
cloud:
config:
uri: http://localhost:8888/configs
You can see user-service replicated to both peer1 and peer2. I can post a PR to your code if you want.
Peer 1
Peer 2
#spencergibb's didn't mention why this hack-ish workaround is required. There is a gotcha with running more than one Eureka server on the same host. Netflix code (com.netflix.eureka.cluster.PeerEurekaNodes.isThisMyUrl) filters out the peer URLs that are on the same host. This may have been done to prevent the server registering as its own peer (I’m guessing here) but because they don’t check for the port, peer awareness doesn’t work unless the Eureka hostnames in the eureka.client.serviceUrl.defaultZone are different. The hacky workaround for this is to define unique hostnames and then map them to 127.0.0.1 in the /etc/hosts file (or its Windows equivalent).
I've created a blog post with the details of Eureka here, that fills in some missing detail from Spring doc or Netflix blog. It is the result of several days of debugging and digging through source code.