Issue with THD::awake and ha_kill_connection

[Raghavendra D Prabhu]

After porting of twitter's timeout patch in
https://blueprints.launchpad.net/percona-server/+spec/port-max-statement-time, it is possible for deadlock to occur when THD::awake is
invoked from within innodb itself. One is the case presented
here: https://bugs.launchpad.net/percona-xtradb-cluster/5.6/+bug/1233301
where the wsrep code calls THD::awake during a BF abort and deadlocks
on lock_sys->mutex and trx->mutex. It is better to fix this in innodb
itself since otherwise it may require changes outside it.

This affects both 5.5 and 5.6 trees.

[Valerii Kravchuk (valerii-kravchuk)]

If this is still relevant for Percona Server, do you have any idea of a test case to run to confirm this?

[Launchpad Janitor (janitor)]

[Expired for Percona Server because there has been no activity for 60 days.]

[Launchpad Janitor (janitor)]

[Expired for Percona Server 5.6 because there has been no activity for 60 days.]

[Launchpad Janitor (janitor)]

[Expired for Percona Server 5.5 because there has been no activity for 60 days.]

[Shahriyar Rzayev (rzayev-sehriyar)]

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PS-3037