Questions tagged with Amazon Aurora

Content language: English

Sort by most recent

Browse through the questions and answers listed below or filter and sort to narrow down your results.

I have an aurora postgresql serverless v2 cluster with 2 instances, one writer and another reader and would like them to scale independently. According to [documentation](https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/aurora-serverless-v2-administration.html#aurora-serverless-v2-choosing-promotion-tier:~:text=Aurora%20Serverless%20v2%20reader%20DB%20instances%20in%20tiers%202%E2%80%9315%20don%27t%20have%20the%20same%20constraint%20on%20their%20minimum%20capacity.%20When%20they%20are%20idle%2C%20they%20can%20scale%20down%20to%20the%20minimum%20Aurora%20capacity%20unit%20(ACU)%20value%20specified%20in%20the%20cluster%27s%20capacity%20range.), if the reader instance is in a Failover priority other than 0 or 1. They SHOULD scale independently but still no matter what I do, they always scale synchronously. I have a workload that runs twice a day and demands the higher acu count but the reader instance has very low usage, so I would like them to scale independently to save on costs. In my use case it is not a problem if there is a longer downtime for the reader instance to scale and take over in case of failure. Thanks
2
answers
0
votes
9
views
asked a day ago
We recently migrated our RDS databases to 5.7+ to prepare for AWS' retirement of MySQL 5.6 support. We have snapshots of previous databases from the 5.6 days - will those be accessible down the line or should we plan to upgrade them? Per the [announcement here](https://repost.aws/questions/QUImshxjRKSRq-t-AQppM6SA/announcement-deprecated-amazon-relational-database-service-rds-for-my-sql-5-6-end-of-life-date-is-august-3-2021): > You can continue to restore your MySQL 5.6 snapshots as well as create read replicas with version 5.6 until the August 3, 2021 end of support date. This makes it sound like we should prepare to restore, upgrade, and re-snapshot existing snapshots to be able to maintain access to them. Is this something Amazon is planning to automate or should I make a ticket for our own teams to do it ourselves?
1
answers
0
votes
20
views
asked 3 days ago
I tried to upgrade from Aurora MySQL 5.7 (2.10.2) to Aurora MySQL 8.0 (3.02.2) and I got this pre-check error in the logs. ``` { "id": "engineMixupCheck", "title": "Tables recognized by InnoDB that belong to a different engine", "status": "OK", "description": "Error: Following tables are recognized by InnoDB engine while the SQL layer believes they belong to a different engine. Such situation may happen when one removes InnoDB table files manually from the disk and creates e.g. a MyISAM table with the same name.\n\nA possible way to solve this situation is to e.g. in case of MyISAM table:\n\n1. Rename the MyISAM table to a temporary name (RENAME TABLE).\n2. Create some dummy InnoDB table (its definition does not need to match), then copy (copy, not move) and rename the dummy .frm and .ibd files to the orphan name using OS file commands.\n3. The orphan table can be then dropped (DROP TABLE), as well as the dummy table.\n4. Finally the MyISAM table can be renamed back to its original name.", "detectedProblems": [ { "level": "Error", "dbObject": "mysql.general_log_backup", "description": "recognized by the InnoDB engine but belongs to CSV" } ] }, ``` Looking at the [MySQL shell code ](https://github.com/mysql/mysql-shell/blob/8.0.23/modules/util/upgrade_check.cc#L1301-L1316) and running that SQL, I get this result. ``` SELECT a.table_schema, a.table_name, concat('recognized by the InnoDB engine but belongs to') FROM information_schema.tables a JOIN (SELECT substring_index(NAME, '/', 1) AS table_schema, substring_index(substring_index(NAME, '/', -1), '#', 1) AS TABLE_NAME FROM information_schema.innodb_sys_tables WHERE NAME like '%/%') b ON a.table_schema = b.table_schema AND a.table_name = b.table_name WHERE a.engine != 'Innodb' +--------------+--------------------+----------------------------------------------------------+ | table_schema | table_name | concat('recognized by the InnoDB engine but belongs to') | +--------------+--------------------+----------------------------------------------------------+ | mysql | general_log_backup | recognized by the InnoDB engine but belongs to | +--------------+--------------------+----------------------------------------------------------+ 1 row in set (0.11 sec) ``` And it is because this entry is present in the information_schema.innodb_sys_tables which should not really be present in the first place. ``` mysql> select * from information_schema.innodb_sys_tables where NAME like '%general%'; +----------+--------------------------+------+--------+-------+-------------+------------+---------------+------------+ | TABLE_ID | NAME | FLAG | N_COLS | SPACE | FILE_FORMAT | ROW_FORMAT | ZIP_PAGE_SIZE | SPACE_TYPE | +----------+--------------------------+------+--------+-------+-------------+------------+---------------+------------+ | 16462 | mysql/general_log_backup | 33 | 9 | 16448 | Barracuda | Dynamic | 0 | Single | +----------+--------------------------+------+--------+-------+-------------+------------+---------------+------------+ 1 row in set (0.09 sec) ``` Coincidentally, according to the release notes of [Aurora 3.02.0](https://docs.aws.amazon.com/AmazonRDS/latest/AuroraMySQLReleaseNotes/AuroraMySQL.Updates.3020.html), it says this: > Fixed an issue that can cause upgrade failures from Aurora MySQL 2 to Aurora MySQL 3 due to schema inconsistency errors reported by upgrade prechecks for the general log and slow log tables. While it says "schema inconsistency errors" and my error is "engineMixupCheck", it feels like both errors are somewhat related to each other since it involves the general_log. Also, when I look at [this](https://repost.aws/questions/QUPC7D-_ZuTgCZSLALluxW9g/need-help-in-upgrading-the-aurora-mysql-5-7-to-mysql-8-urgent), it mentions about > mysql.general_log_backup recognized by the InnoDB engine but belongs to CSV. which is exactly the error that I am getting but it does not seem a solution has been provided. So, has anyone seen this error and is there a workaround for this?
0
answers
0
votes
62
views
asked 3 days ago
It appears that Aurora PostgreSQL major version upgrade requires us to first drop all replication slots, then perform upgrade, and then recreate replication slots. We use logical slots for Debezium/Kafka replication for inter-process work flows. When we drop and recreate these replication slots as part of major version upgrade, how can we ensure replication restarts from where it left of (meaning, replication resumes incremental changes only) and not force us to do FULL sync. We cannot afford to have FULL sync due to large table sizes.
1
answers
0
votes
23
views
asked 3 days ago
Our instance with a single reader/writer had been humming along for some time without issue. Then last week in the middle of the night it got stuck in a bit of an automatic recovery loop, going through the full recovery process about 6 times. It did the same thing again this morning. I know we are responsible for managing these sorts of outages, and plan to add a second instance per AWS recommendation and recommendations here on similar threads, but this seems a bit abnormal? A single recovery due to bad hardware or an underlying systems change is one thing, getting stuck in a loop for several hours, recovering multiple times, does not? The main messages to kick it off is: "Recovery of the DB instance has started. Recovery time will vary with the amount of data to be recovered." We don't see any issues in the underlying MySql logs, just repeated startups, the recovery happens in the middle of the night (us-west-2), and we haven't made any recent changes except bumping to 5.7 several weeks ago.
2
answers
0
votes
19
views
asked 5 days ago
I have a Aurora MySQL db.r5large instance in us-east1 region and the 'Add region' operation with us-east2 as the secondary region fails to create instances for the secondary cluster with instance creation stuck with 'Creating' state for more than 24 hours. There is absolutely no indication of what is wrong.
1
answers
0
votes
5
views
asked 5 days ago
Are IAM roles (associated with the Blue cluster) replicated to the Green one during the Blue/Green environment creation?
1
answers
0
votes
31
views
AWS
asked 9 days ago
Hi there, I often seen it mention in the documentation that you can encrypt an Aurora Read-Replica, but since all instances share the same underlying storage and data is not encrypted in memory then what is the point of it? Thanks in advance.
1
answers
0
votes
17
views
asked 12 days ago
Hi, I was going through the below reinvent video about Deep dive on Amazon Aurora with PostgreSQL. I see a mention about "Concurrency : Remove Log buffer" and "Aurora PostgreSQL: Writing Less". So does this mean that Aurora Postgres doesn't use wal buffer or is there is there any change in the way it is being used? https://www.youtube.com/watch?v=Ul-j5fKfv2k&t=334s Thanks,
1
answers
0
votes
25
views
Rakesh
asked 12 days ago
Setting a custom value works fine in Aurora. For example: ``` SET jwt.claims.email = 'user.email@example.com'; SET my.custom.setting = 'yes'; ``` In vanilla Postgres, one can also associate custom settings with a database object: ``` ALTER DATABASE mydb SET my.custom.setting = 'yes'; ``` But in Aurora, I get the response: ``` ERROR: permission denied to set parameter "my.custom.setting"; SQLState: 42501 ``` even though the current user is the database owner. Why does this disparity between standard Postgres and Aurora (and presumably RDS) exist? This breaks compatibility with a current setup that we're trying to migrate to AWS.
0
answers
0
votes
20
views
asked 17 days ago
I'm trying to set the a default value for the parameter `aurora_replica_read_consistency` as in the docs https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/aurora-global-database-write-forwarding.html#aurora-global-database-write-forwarding-params The parameter is missing from DB cluster parameters or DB parameters and it seems cannot be added manually. Is there a way to set a default value? Please note that setting this parameter for every session is not an option.
1
answers
0
votes
21
views
giupas
asked 18 days ago
Hello Following the aws link https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/CHAP_Troubleshooting.html we decided to enable the aurora_oom_response parameter with the value print, tune, decline, kill_query in our DB clusters' parameter groups but we have noticed that the value appears as empty when using the mysql client: ``` mysql> show global variables like '%oom%'; +---------------------+-------+ | Variable_name | Value | +---------------------+-------+ | aurora_oom_response | | +---------------------+-------+ 1 row in set (0.00 sec) mysql> ``` We even tried rebooting the instances but the value still shows as empty in mysql client although is clearly setup in the right Parameter groups (cluster and instance). Is this a bug in Aurora? the engine used is 5.7 version 2.11 Thanks!
0
answers
0
votes
32
views
Kr4t0s
asked 22 days ago