error upgrading tdengine from 3.3.6.0 to 3.3.6.3 in kubernetes

k8s版本:
Client Version: v1.31.0
Kustomize Version: v5.4.2
Server Version: v1.31.7-eks-bcf3d70
helm chart :3.0.2
实践记录:
1 先升级后写入数据(没有问题)
在跑3.3.6.0的镜像时,会出现正常情况pod内进程如下

更换成3.3.6.3之后:

将镜像版本由td-engine由3.3.6.0升级到3.3.6.3(之前没有写入数据),就不会报错:日志如下:
pod也是健康的:

执行如下命令写入数据也没问题
taosBenchmark -d db -t 100 -n 1000 -T 4 -I stmt -y
2 先写入数据后升级(出现问题)
先准备个tdengine3.3.6.0 3节点集群

然后写入数据:
taosBenchmark -d db -t 100 -n 1000 -T 4 -I stmt -y

再改镜像:

pod滚动更新,接下来会看到日志报错:
pod-2:

进程都正常:

pod-1:

taoskeeper没起来:

pod-0:

taoskeeper没起来:

3 直接将镜像改成3.3.6.3 (出现问题)

将镜像直接变成3.6.6.3,镜像直接起不来

Reproduce the upgrade sequence with Kubernetes v1.31, Helm chart 3.0.2, and TDengine images 3.3.6.0 and 3.3.6.3. Start with the taoskeeper and pod logs while running taosBenchmark, comparing upgrades before and after data is written. Done means a three-node cluster with existing data upgrades successfully and all pods remain healthy.