-
Notifications
You must be signed in to change notification settings - Fork 66
Ingest failed due to EpochNotMatch error #436
Comments
This error means raft member has changed. (conf_ver: raft member version, version: region range version). More errors from lightning log, all of them are conf_ver mismatch. Does lightning local backend wait scatter? @glorv More errors
Also from tikv log, lightning batch splits TiKV 2 times, while TiKV itself splits about 683 times, it may be caused by ingest multiple sst file pairs(write/default) in one region. We should ensure one sst file pair one regoin.
|
This is misleading, it's not the root cause.
The actual error causes lightning failure is From tikv log, I find region 8348 has been ingested multiple times, and before every ingests, it splits. 8348 ingest split
|
There are three kinds of event, that can cause
|
Since the
Further: |
We have optimized the stability of lightning local backend in the past few months:
|
Bug Report
Please answer these questions before submitting your issue. Thanks!
What did you do? If possible, provide a recipe for reproducing the error.
import with lightning local backend
What did you expect to see?
import successfully
What did you see instead?
Versions of the cluster
TiDB-Lightning version (run
tidb-lightning -V
):TiKV version (run
tikv-server -V
):Operation logs
Configuration of the cluster and the task
tidb-lightning.toml
for TiDB-Lightning if possibletikv-importer.toml
for TiKV-Importer if possibleinventory.ini
if deployed by AnsibleScreenshot/exported-PDF of Grafana dashboard or metrics' graph in Prometheus for TiDB-Lightning if possible
The text was updated successfully, but these errors were encountered: