we are working only with flink in our team, and we...
# troubleshooting
u
we are working only with flink in our team, and we having a lot of troubles for creating real stable flink jobs, all of our workloads restart or crash once a day or week. We did a deep research and we know its because of checkpoints or rocksdb (and mainly the conbination of the 2 of them). We are using flink 1.19 and flink 1.16 do you know if its getting solved in newer versions?
a
We ran Flink 1.19 for a long time (now we're using 1.20) with checkpointing and RocksDB without any issues related to checkpointing. Do you know what specifically causes your job restarts and crashes? Flink 1.19 is still being maintained. 1.19.3 was released today. Are you using the latest 1.19?
p
I ran 200+ Flink deployments among different clusters and with quite some load. They were pretty stable. We even had high-priority/operational features fully reliant on Flink. Perhaps there is something else jeopardizing your jobs.
u
Hello my friends, Does anyone know where i can find the complete and full flink 1.16 java 11 dockerfile?