Lily2025

Results 106 comments of Lily2025

pitr lag reached 28mins when tikv rolling restart

lag is 10~16 mins now

pitr lag reached 28mins when tikv rolling restart

/remove-type bug /type enhancement

enable stale read and closest-replicas，qps of all az drop to bottom during simulating network partition between two az

/severity major

enable stale read and closest-replicas，qps of all az drop to bottom during simulating network partition between two az

/assign cfzjywxk

pitr status became paused after restart all pd、tikv、tidb

/type bug /severity major /assign Leavrth

other tikv panic when inject one of tikv failure with ticdc changfeed running

/type bug /severity critical /assign hicqu

tikv always report “Region error (will back off and retry)” when disk capacity is near full with follower read

/assign zyguan

tikv always report “Region error (will back off and retry)” when disk capacity is near full with follower read

/assign cfzjywxk

tikv always report “Region error (will back off and retry)” when disk capacity is near full with follower read

from @MyonKeminta almost full nodes (all followers but no leader) refuse to apply new logs from the leader. As a result, the region epoch is not updated when it changes....

QPS fell to zero for a few minutes after fault recover from minority tikv network-loss due to full of "Async Apply CPU"

qps drop after one of tikv recover from io hang ![image](https://github.com/tikv/tikv/assets/84712107/42a6776f-4d2c-431b-8e50-a43c9d1130b8) ![image](https://github.com/tikv/tikv/assets/84712107/d5bc2d5f-ec47-4928-914b-e79bf9ff04bb) https://clinic.pingcap.com.cn/portal/#/orgs/31/clusters/7319161375831017483?from=1704139208&to=1704143700 ![image](https://github.com/tikv/tikv/assets/84712107/8cc09b54-84a4-4174-b987-72486ece44b5) ![image](https://github.com/tikv/tikv/assets/84712107/3a133710-05a1-4e26-bc77-049cffc1d1ff) ![image](https://github.com/tikv/tikv/assets/84712107/6f5f43eb-dab7-4e27-b035-8379b8385edf)