0

I'm trying to import a large (30GB) SQL file (generated using mysqldump) into an RDS instance MySQL 5.7 database, from an EC2 instance on the same VPC.

It's importing most without a problem, but missing around 5 tables from the end of the dump.

Looking around for related logs in the AWS console I can't really find much of use, all I'm seeing is this:

2021-02-17T12:16:20.168722Z 0 [Note] InnoDB: page_cleaner: 1000ms intended loop took 10360ms. The settings might not be optimal. (flushed=599 and evicted=179, during the time.)
2021-02-17T12:16:36.461195Z 0 [Note] InnoDB: page_cleaner: 1000ms intended loop took 5988ms. The settings might not be optimal. (flushed=490 and evicted=124, during the time.)
2021-02-17T12:16:46.848239Z 0 [Note] InnoDB: page_cleaner: 1000ms intended loop took 6766ms. The settings might not be optimal. (flushed=447 and evicted=0, during the time.)
2021-02-17T12:17:16.371538Z 0 [Note] InnoDB: page_cleaner: 1000ms intended loop took 10380ms. The settings might not be optimal. (flushed=370 and evicted=240, during the time.)
2021-02-17T12:19:22.153786Z 0 [Note] InnoDB: page_cleaner: 1000ms intended loop took 27942ms. The settings might not be optimal. (flushed=299 and evicted=637, during the time.)
2021-02-17T12:21:19.368279Z 0 [Note] InnoDB: page_cleaner: 1000ms intended loop took 16509ms. The settings might not be optimal. (flushed=361 and evicted=136, during the time.)
2021-02-17T12:23:01.420623Z 0 [Note] InnoDB: page_cleaner: 1000ms intended loop took 5097ms. The settings might not be optimal. (flushed=349 and evicted=125, during the time.)
2021-02-17T12:23:16.836488Z 0 [Note] InnoDB: page_cleaner: 1000ms intended loop took 8285ms. The settings might not be optimal. (flushed=307 and evicted=121, during the time.)
2021-02-17T12:23:52.965471Z 0 [Note] InnoDB: page_cleaner: 1000ms intended loop took 19104ms. The settings might not be optimal. (flushed=299 and evicted=607, during the time.)
----------------------- END OF LOG ----------------------

I'm not really sure where to start looking without any logs. Only thing I can think of is maybe lack of memory, but looking at the stats it still had "some" memory left on the RDS instance while it was importing.

When I check the file the CREATE TABLE statements are definitely there for the missing tables.

Can anyone give any tips on how I can figure out what's going wrong?

1 Answer 1

0

Turned out this was just lack of memory on the chosen RDS instance. Increased it to the "medium" instance instead of "micro" and it imported fine at around 10x the speed, with all tables.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.