Closed hendrikheil closed 3 weeks ago
I got an issue with events logs clickhouse table and it might be because of this issue
Hey, Do you know if this is going to be resolved soon? For now, I already had to setup this : https://github.com/jitsucom/jitsu/issues/1125#issuecomment-2351757178 and I think I still got issues with logs max size and some other bulker issues
postgresql:
enabled: false
redis:
enabled: false
# For Kafka rotor:
mongodb:
enabled: true
kafka:
extraConfig: |
offsets.topic.replication.factor=1
transaction.state.log.replication.factor=1
controller:
replicaCount: 1
# Clickhouse is now used for Live events:
# https://github.com/jitsucom/jitsu/releases/tag/jitsu2-v2.5.0
clickhouse:
enabled: true
shards: 1
replicaCount: 1
zookeeper:
enabled: false
ingress:
enabled: true
className: "nginx"
annotations:
nginx.ingress.kubernetes.io/enable-access-log: "true"
kubernetes.io/tls-acme: "true"
cert-manager.io/cluster-issuer: letsencrypt-prod
host: <self-hosted>.<domain>.com
tls: true
config:
enabled: true
# database in which we keep jitsu settings:
databaseUrl: "postgres://newjitsu:<password>@<host>:<port>/newjitsu?schema=newjitsu"
# cickhouse internal database, since of we got these issues:
# - https://github.com/jitsucom/jitsu/issues/1125
# - https://github.com/jitsucom/jitsu/issues/1119
clickhouseDatabase: "newjitsu_metrics"
console:
config:
# We use internal custom user:
seedUserEmail: "<USERNAME>"
seedUserPassword: "<PASSWORD>"
disableSignup: "true"
ingress:
enabled: false # because ingress.enabled above
ingest:
ingress:
enabled: false # because ingress.enabled above
Hey, Do you know if this is going to be resolved soon? For now, I already had to setup this : jitsucom/jitsu#1125 (comment) and I think I still got issues with logs max size and some other bulker issues
@stouch Hi, I'm planning on working on this today or tomorrow depending on what time allows. Am I understanding you correctly that the issues you were experiencing initially were solved by setting the Clickhouse database to newjitsu_metrics
? Sounds like this value is assumed some places in the Jitsu code, so if that's the case I'll have to change the default, even if it's potentially breaking.
Bulker issues I imagine should be out of the picture once I've brought things up to speed with v2.8.x as I understand the bulker is no longer needed there.
Hey, Do you know if this is going to be resolved soon? For now, I already had to setup this : jitsucom/jitsu#1125 (comment) and I think I still got issues with logs max size and some other bulker issues
@stouch Hi, I'm planning on working on this today or tomorrow depending on what time allows. Am I understanding you correctly that the issues you were experiencing initially were solved by setting the Clickhouse database to
newjitsu_metrics
? Sounds like this value is assumed some places in the Jitsu code, so if that's the case I'll have to change the default, even if it's potentially breaking.Bulker issues I imagine should be out of the picture once I've brought things up to speed with v2.8.x as I understand the bulker is no longer needed there.
Y'es it changed in the code I explained where it is in this comment : https://github.com/jitsucom/jitsu/issues/1125#issuecomment-2351757178
Oh great if the bulker issues would be resolved 🥳
@stouch Having started looking at this, it seems the bulker isn't going away, rather not depended on by syncctl or its sidecar. What problems are you having with the bulker? If it's something caused by the configuration in the chart it'd be great to be able to fix it for the upcoming release.
Jitsu just released v2.8.0.
Notable: syncctl no longer relies on bulker and instead has bulker embedded into the image directly.