Here are the tables that I have compressed on frdb2001. We should run this compaction on the other hosts (except the current master) in time. Then, when we are ready, we should do a master swap and run the compaction there.
One host we will want to watch the effects on is frdb2002. Since it is the middle master, we should see if the query passes on to frdb2001 (I believe it will). Mostly it's just to be aware of knock on effects.
Hosts to run compaction on:
[x] frdb1001
[ ] frdb1002
[ ] frdb1003
[ ] frdev1001
[x] frdb2001
[x] frdb2002
```
# Tables to compress (increasing time to run)
#
# 18s 5G
compact_innodb_table -d civicrm -t civicrm_entity_tag
# 2m 52G
compact_innodb_table -d drupal -t queue2civicrm_log
# 5m 2G
compact_innodb_table -d civicrm -t civicrm_email
# 15m 4G
compact_innodb_table -d civicrm -t civicrm_address
# 23m 7G
compact_innodb_table -d civicrm -t civicrm_activity_contact
# 40m 13G
compact_innodb_table -d civicrm -t civicrm_contact
# 84m 5G
compact_innodb_table -d civicrm -t civicrm_activity
# 129m 39G
compact_innodb_table -d civicrm -t civicrm_mailing_provider_data
```