Yury Bushmelev
Yury Bushmelev
According to what I saw here, I'd suggest you to kill kafka-backup process by `kill -9` few times. I guess you may get to the state :) I really hope...
Hit into this again after few hours of #88 and one OOM..
Saw this tonight on kafka-backup service shutdown before doing Azure blobstore backup: ``` May 30 19:19:24 backupmgrp1 docker/kafka-backup-chrono_prod[16472]: [2020-05-30 19:19:24,572] INFO WorkerSinkTask{id=chrono_prod-backup-sink-0} Committing offsets synchronously using sequence number 2782: {xxx-4=OffsetAndMetadata{offset=911115,...
Fortunately resolved automagically after restart! 🤞
JFYI, I hit this again.. monitoring helps to catch this early but anyway I should do restart manually :(
It was just lag of backup sink consumer group.. I still had no chance to inject jmx_exporter yet
Got it again :(
Today found it with a bit more info. It happens after coordinator change in kafka cluster: ``` [2020-06-25 04:10:54,718] INFO [AdminClient clientId=adminclient-2] Node kafka2.tld:9093 (id: 2 rack: null) is no...
I realised I need this (as I understand retention). Currently we're uploading everything into Azure blobstore nightly. Nothing is deleted from blobstore so far. I just found if we will...
> * Copy only the data/segments you need (or delete the ones you do not need) So last question is how to find do I need this particular segment or...