0

I am experiencing a strange issue: I have a t2.micro AWS RDS instance, but it constantly eat its free space. The current usage of the tables, checked with this query: 

 select table_schema, sum((data_length+index_length)/1024/1024) AS MB from information_schema.tables group by 1

result:

# TABLE_SCHEMA, MB
mysql, 5.89062500
information_schema, 0.00000000
sys, 0.01562500
performance_schema, 0.00000000
testdb_v6.2, 0.81250000

Please check the attached picture. disk usage
When I experienced the out of space issue, I increased the available space by 2GB, but it constantly eat it. Can you suggest how to get through this? I checked that there is no data consumed by any kind of service, everything looks normal.

Dave M
  • 4,514
  • 22
  • 31
  • 30
mkiss
  • 103
  • 1
  • Okay, temporary solution: a restart solved a problem, but I had to do twice cause first was not enough. Keeping update this thread if I found a permanent solution. – mkiss Dec 14 '19 at 19:28
  • That's a rotten graph -- no units for y axis. Apparently no readings for 3 days, falsely implying a steady decline of space. – Rick James Dec 14 '19 at 21:51
  • The Y axis in unit MB, this is defined in the dropdown list. – mkiss Dec 15 '19 at 11:50
  • What is the setting if `innodb_buffer_pool_size`? – Rick James Dec 15 '19 at 16:01
  • `{DBInstanceClassMemory*3/4}` the instance has 1GB memory. "Returns the number of bytes of memory allocated to the DB instance class associated with the current DB instance, less the memory used by the Amazon RDS processes that manage the instance." – mkiss Dec 15 '19 at 16:12

1 Answers1

0

3/4*mem is a handy formula for larger VMs. But it is dangerously small for a 1GB VM (or 1GB real machine). It could lead to swapping, which is terrible for performance. And, since you have only a few MB of data, I suggest only 128M.

Hopefully, that change will stop the symptom you are seeing.

Rick James
  • 2,463
  • 1
  • 6
  • 13