'ibf_core_store' bottleneck causing server crash

[4UM_+min]

Hi,

It is our peak traffic season and we are running into an issue with queries timing out that eventually crash the server. It happens several times a day.

/*IPS\Data\Store\_Database::set:89*/ REPLACE INTO `ibf_core_store` ( `store_key`, `store_value` ) VALUES ( ?, ? ) 

IPS\Db\Exception: Lock wait timeout exceeded; try restarting transaction (1205); 

We didn't have this issue last year under the same traffic load, so it seems something in our server configuration or software has changed. The table only has 109 rows in it, but takes a very long time to update.

Our server load is still very low during these crashes, so it seems like there is a bottleneck with this table.

Any ideas? I've seen others mention making sure memcached was enabled and running, which appears to be, but I'm triple checking.

[bfarber]

You may wish to switch to Redis, which will use Redis for both caching and data store.

You can also disable using the database for the datastore and store files on disk instead, which may work out better for you.