Aurora MySQL Upgrade From 2.10.1 to 3.01.0 (MySQL 5.7 to 8) failing

1

I have two clusters I would like to upgrade. I am restoring from snapshot. After creating the new cluster it falls back to V2, I have 5 errors in the precheck log:

Aurora Fast Online DDL is not supported from 3.x onwards. This check verifies that the database do not have artifacts realted to the feature

Table information_schema.INNODB_SYS_INDEXES_HISTORY is not empty. Your database has pending Aurora Fast online DDL operations. Reachout to AWS support for assistance

Table information_schema.INNODB_SYS_FIELDS_HISTORY is not empty. Your database has pending Aurora Fast online DDL operations. Reachout to AWS support for assistance

Table information_schema.INNODB_SYS_SCHEMA_HISTORY is not empty. Your database has pending Aurora Fast online DDL operations. Reachout to AWS support for assistance

Table information_schema.INNODB_SYS_COLUMNS_HISTORY is not empty. Your database has pending Aurora Fast online DDL operations. Reachout to AWS support for assistance

Table information_schema.INNODB_SYS_TABLES_HISTORY is not empty. Your database has pending Aurora Fast online DDL operations. Reachout to AWS support for assistance

I had turned labmode on both clusters. Meanwhile the serer

demandé il y a 2 ans1935 vues
1 réponse
1

Hi,

I understand you faced precheck errors when trying to upgrade Aurora version from 2.x to 3.x. The errors messages you shared here is related to known issue that due to pending DDL updates corresponding to Fast DDL operations listed below. [+] https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/AuroraMySQL.Managing.FastDDL.html#AuroraMySQL.Managing.FastDDL-v2This feature is depreciated in all Mysql 8.0 compatible Aurora versions. Hence, it is necessary to apply these pending DDL updates to be able to upgrade to 3.<> version.

Please try to run optimize table command on all the tables in 5.7 source cluster (on which DDL queries were executed) and then took the snapshot again. Running optimize table will basically empty all the 5 history tables mentioned above which were erroring out. You can then restore the snapshot to aurora mysql 8.0.

Note: I would suggest to test on a test environment by restoring a cluster from 5.7 snapshot and run commands before applying to production environment.

mysql>optimize table <table name>;

[+] https://dev.mysql.com/doc/refman/5.7/en/optimize-table.html

AWS
INGÉNIEUR EN ASSISTANCE TECHNIQUE
Kevin_Z
répondu il y a 2 ans

Vous n'êtes pas connecté. Se connecter pour publier une réponse.

Une bonne réponse répond clairement à la question, contient des commentaires constructifs et encourage le développement professionnel de la personne qui pose la question.

Instructions pour répondre aux questions