Logs:
[root@lvndev012336 kafka-aiops]# kubectl logs pod/cluster1-proxysql-0 -c proxysql -n do-kafka-v20 |grep error
[root@lvndev012336 kafka-aiops]# kubectl logs pod/cluster1-proxysql-1 -c proxysql -n do-kafka-v20 |grep error
2021-03-04 20:50:07 MySQL_Monitor.cpp:1845:monitor_galera_thread(): [ERROR] Got error. mmsd 0x7f4f3e0259c0 , MYSQL 0x7f4f3ba26e00 , FD 27 : timeout check
[root@lvndev012336 kafka-aiops]# kubectl logs pod/cluster1-proxysql-2 -c proxysql -n do-kafka-v20 |grep error
2021-03-04 20:50:07 MySQL_Monitor.cpp:1845:monitor_galera_thread(): [ERROR] Got error. mmsd 0x7f3816231400 , MYSQL 0x7f3814a10000 , FD 28 : timeout check
Hello @saloni24 ,
so what happened here?
Could you please share:
- version of the operator you use
- your cr.yaml
- steps to reproduce this issue
I am too getting this error.
Here is my CR:
apiVersion: pxc.percona.com/v1
kind: PerconaXtraDBCluster
metadata:
finalizers:
- delete-pxc-pods-in-order
- delete-proxysql-pvc
- delete-pxc-pvc
generation: 1
name: minidatapool
namespace: ec-core
spec:
allowUnsafeConfigurations: true
backup:
image: percona-xtradb-cluster-operator:1.7.0-pxc8.0-backup
pitr:
enabled: false
schedule:
- keep: 5
name: monthly-backup
schedule: 0 0 1 * *
storageName: fs-pvc
serviceAccountName: minidatapool-sa
storages:
fs-pvc:
s3: null
type: filesystem
volume:
emptyDir: null
hostPath: null
persistentVolumeClaim:
accessModes:
- ReadWriteOnce
resources:
requests:
storage: 10Gi
haproxy:
affinity:
advanced: null
antiAffinityTopologyKey: none
enabled: false
gracePeriod: 30
image: percona-xtradb-cluster-operator:1.7.0-haproxy
podDisruptionBudget:
maxUnavailable: 1
resources:
requests:
cpu: 200m
memory: 512Mi
serviceAccountName: minidatapool-sa
size: 1
logcollector:
enabled: true
image: percona-xtradb-cluster-operator:1.7.0-logcollector
pmm:
enabled: false
image: pmm-client:2.12.0
proxysql:
affinity:
advanced: null
antiAffinityTopologyKey: none
enabled: true
gracePeriod: 30
image: percona-xtradb-cluster-operator:1.7.0-proxysql
podDisruptionBudget:
maxUnavailable: 1
resources:
requests:
cpu: 200m
memory: 512Mi
serviceAccountName: minidatapool-sa
size: 1
volumeSpec:
emptyDir: null
hostPath: null
persistentVolumeClaim:
resources:
requests:
storage: 10Gi
pxc:
affinity:
advanced: null
antiAffinityTopologyKey: none
enabled: true
gracePeriod: 600
image: percona-xtradb-cluster:8.0.21-12.1
podDisruptionBudget:
maxUnavailable: 1
resources:
requests:
cpu: 200m
memory: 512Mi
serviceAccountName: minidatapool-sa
size: 1
volumeSpec:
emptyDir: null
hostPath: null
persistentVolumeClaim:
resources:
requests:
storage: 10Gi
secretsName: minidatapool-secret
sslInternalSecretName: my-cluster-ssl-internal
sslSecretName: my-cluster-ssl
I have created the users following this link:
I run into multiple issues when trying to connect to a database using one of the users created using above step. Sometimes I get access denined error when conecting using proxysql and sometimes I get the error reported in this thread.
However if I connect to pxc directy it works.
Is there a way to connext to proxysql pod and get more details on why it is not working or if users are getting synced to proxysql or not. Right now I am not able to connect to proxysql or I am not able to figure out how.
I tried following this document but it just not able to login to the mysql instance there to run the desired queries.
https://www.percona.com/doc/percona-xtradb-cluster/LATEST/howtos/proxysql-v2.html
This is the additional logs from proxysql.
2021-05-27 06:39:01 MySQL_Session.cpp:6600:unable_to_parse_set_statement(): [WARNING] Unable to parse unknown SET query from client 10.7.4.21:37516. Setting lock_hostgroup. Please report a bug for future enhancements:SET foreign_key_checks=1, sql_safe_updates=0
2021-05-27 06:43:18 MySQL_Session.cpp:3941:handler(): [WARNING] Error during query on (10,minidatapool-pxc-0.minidatapool-pxc.ec-core.svc.cluster.local,3306,1632): 1142, SELECT command denied to user 'managerdb_9a12a062_a'@'minidatapool-proxysql-0.minidatapool-proxysql-unready.ec-core.sv' for table 'user_variables_by_thread'
I’m get this error too.
I think it is related to this bug: [K8SPXC-534] No servers in hostgroup 10 during pxc statefulset update - Percona JIRA
It is going to be resolved in 1.11 version of the Operator.