MySQL Router Bootstrap Configuration on Windows 10 64bit fails not identifying Cluster Metadata

User_73TNT

I am trying to configure MySQL Router on Windows 10 64 Bits, for  monitoring an InnoDB Cluster 8.0.19 that is running on 3 Linux VMs.  But when configuring the Windows MySQL Router, I receive the following error:

Beginning configuration step: Creating MySQL Router configuration files

An Error occurred: Error: Expected the metadata server to contain configuration for one cluster, found none.

See https://dev.mysql.com/doc/refman/8.0/en/mysql-innodb-cluster-creating.html about how to create a cluster.

Ended configuration step: Creating MySQL Router configuration files

My Testing environment for InnoDB Cluster is on 3 Linux VMs.  Here is the Status Result on the Cluster.

MySQL  vm-mysqlee2:3306 ssl  JS > cluster.status()

{

    "clusterName": "mysqlee_cluster",

    "defaultReplicaSet": {

        "name": "default",

        "primary": "vm-mysqlee2.ita.local:3306",

        "ssl": "REQUIRED",

        "status": "OK",

        "statusText": "Cluster is ONLINE and can tolerate up to ONE failure.",

        "topology": {

            "vm-mysqlee1.ita.local:3306": {

                "address": "vm-mysqlee1.ita.local:3306",

                "mode": "R/O",

                "readReplicas": {},

                "replicationLag": null,

                "role": "HA",

                "status": "ONLINE",

                "version": "8.0.19"

            },

            "vm-mysqlee2.ita.local:3306": {

                "address": "vm-mysqlee2.ita.local:3306",

                "mode": "R/W",

                "readReplicas": {},

                "replicationLag": null,

                "role": "HA",

                "status": "ONLINE",

                "version": "8.0.19"

            },

            "vm-mysqlee3.ita.local:3306": {

                "address": "vm-mysqlee3.ita.local:3306",

                "mode": "R/O",

                "readReplicas": {},

                "replicationLag": null,

                "role": "HA",

                "status": "ONLINE",

                "version": "8.0.19"

            }

        },

        "topologyMode": "Single-Primary"

    },

    "groupInformationSourceMember": "vm-mysqlee2.ita.local:3306"

}

I have disabled IPv6 on Linux and Windows Machines.  Someone have experienced this error with the Windows MySQL Router Configuration.?

Regards,

lefred-MySQL-Oracle

Hi,

This is normal some changes were done in MySQL InnoDB Cluster. As you are using a newer version of 8.0.21 (which is often recommended), You didn't upgrade with the Shell the metadata from your the previous metadata created before 8.0.19 (https://mysqlserverteam.com/upgrading-mysql-innodb-cluster-metadata/

Cheers,

Dave Stokes-MySQL Community Team-Oracle

What did you router bootstrap command look like?

User_73TNT

First I used the MySQL Installer GUI where it first validate the credentials to a mysql node, and shows you the ports that it will be using.  Then when this is step is ok, the next step where you execute the configuration it send the message.

Then I tried with the following command line and get the same message.

         C:\Program Files\MySQL\MySQL Router 8.0\bin>.\mysqlrouter.exe --bootstrap [email protected]:3306 -d C:\MySQL-Router

Regards,

Regards,

lefred-MySQL-Oracle

Hi,

This is normal some changes were done in MySQL InnoDB Cluster. As you are using a newer version of 8.0.21 (which is often recommended), You didn't upgrade with the Shell the metadata from your the previous metadata created before 8.0.19 (https://mysqlserverteam.com/upgrading-mysql-innodb-cluster-metadata/

Cheers,

User_73TNT

Before I tryied with a Community Edition Router 8.0.19, and received the same error.  Does it is also normal, when using a Community Router with a Commercial InnoDB Cluster..?

On the eDeliviry I didn't find how to download eh 8.0.19 archive/version, just the 8.0.21.. does it is available for downloading..?

Regards,

lefred-MySQL-Oracle

This change happened at 8.0.19, please just upgrade the metadata:

Connect with MySQL Shell to the Primary node and just do: dba.upgradeMetadata()

Then bootstrap router again.

Cheers,

User_73TNT

Thanks for the advice, and I understand that it should be done, but I have to share with you the following.

Since all my environment is 8.0.19, I have configured before other router from Linux Box without problem.

Comming back to my issue, with your comments, I proceeded to  enable the General Log of the Instance where bootstraping is connecting and found this queries:

    2020-07-25T01:24:45.271404Z 12612 Connect [email protected] on  using SSL/TLS

    2020-07-25T01:24:45.271771Z 12612 Query SELECT * FROM mysql_innodb_cluster_metadata.schema_version

    2020-07-25T01:24:45.272285Z 12612 Query select cluster_type from mysql_innodb_cluster_metadata.v2_this_instance

    2020-07-25T01:24:45.273474Z 12612 Query select ((select count(*) from mysql_innodb_cluster_metadata.v2_gr_clusters)=1) as has_one_gr_cluster

So I executed this queries on the instances with the following results:

   

The last query returns a cero (o).  Then I verified that this last table mysql_innodb_cluster_metadata.v2_gr_clusters has two (2) records instead of one (1) .

  

I think, this was the really reason of the message I was receiving.  In addition I found the following Doc on My Oracle Support:

    MySQL Router Bootstrap fails With Error: "Error: The provided server contains an unsupported cluster metadata." (Doc ID 2686089.1)

When applying the solution, to drop/delete the old record, I had an issue. but it is something that is not related to my original question.

I hope this help any who may be facing a similar issue.

Regards,