Data integrity on Analytics Kafka nodes
Closed, ResolvedPublic
Actions

Assigned To

Authored By

	elukey
	Feb 3 2016, 11:22 AM

Description

Context https://wikitech.wikimedia.org/wiki/Analytics/Cluster/Kafka/Administration#Data_integrity

A potential fix could be to investigate the possibility of using RAID even for broker's partition logs and configure it during the next cluster installation (for example, for the Kafka 0.9 migration).

Short term fixes might include having SMART icinga alerts on our IRC Channel.

Related Objects

Mentioned Here: T99105: Kafka Broker disk usage is imbalanced

Event Timeline

elukey created this task.Feb 3 2016, 11:22 AM

elukey claimed this task.

elukey raised the priority of this task from to Needs Triage.

elukey updated the task description. (Show Details)

elukey added a project: Analytics.

elukey subscribed.

Restricted Application added subscribers: StudiesWorld, Aklapper. · View Herald TranscriptFeb 3 2016, 11:22 AM

Kind of the same as T99105, if we move to RAID that'll probably solve this

Data integrity on Analytics Kafka nodesClosed, ResolvedPublicActions

Description

Related Objects

Event Timeline

Data integrity on Analytics Kafka nodes
Closed, ResolvedPublic
Actions