ha trove service can't connect to mysql server once in a while

Bug #1505977 reported by yang on 2015-10-14
This bug affects 3 people
Affects Status Importance Assigned to Milestone
OpenStack DBaaS (Trove)
Fix Released
Matthew Van Dijk

Bug Description

I have installed openstack HA juno , referencing

Trove service can not connect to mysql server once in a while.
Error logs in trove-api.log:

2015-10-14 16:24:28.332 6608 TRACE trove.common.wsgi OperationalError: (OperationalError) (2006, 'MySQL server has gone away') 'SELECT instances.id AS instances_id, instances.created AS instances_created, instances.updated AS instances_updated, instances.name AS instances_name, instances.hostname AS instances_hostname, instances.compute_instance_id AS instances_compute_instance_id, instances.task_id AS instances_task_id, instances.task_description AS instances_task_description, instances.task_start_time AS instances_task_start_time, instances.volume_id AS instances_volume_id, instances.flavor_id AS instances_flavor_id, instances.volume_size AS instances_volume_size, instances.tenant_id AS instances_tenant_id, instances.server_status AS instances_server_status, instances.deleted AS instances_deleted, instances.deleted_at AS instances_deleted_at, instances.datastore_version_id AS instances_datastore_version_id, instances.configuration_id AS instances_configuration_id, instances.slave_of_id AS instances_slave_of_id, instances.cluster_id AS instances_cluster_id, instances.shard_id AS instances_shard_id, instances.type AS instances_type \nFROM instances \nWHERE instances.deleted = 0 AND instances.tenant_id = %s AND instances.cluster_id IS NULL ORDER BY instances.id \n LIMIT %s' (u'56b645d728c0484fa2d445fe28800c32', 21)

Services like nova, keystone,heat can connect to mysql server well.

Jeremy Stanley (fungi) wrote :

I don't see anything about this bug report which looks like a security vulnerability, and suspect it was intended to be opened as a normal bug instead. Can you clarify why this is set to "private security" rather than just "public"?

Craig Vyvial (cp16net) wrote :

I see nothing as well that makes me think this is a security issue.

yang (yangwang1211) on 2015-10-15
information type: Private Security → Public
György Szombathelyi (gyurco) wrote :

I also encountering with this bug, I see trove uses an ancient SQLAlchemy, but other OS components are moved to oslo_db with more recent SQLAlchemy, which can reconnect to the DB in case of terminated connection by the MySQL server.

Amrith Kumar (amrith) wrote :

There is also a blueprint for this


I'm going to mark this bug appropriately.

Changed in trove:
status: New → Confirmed
milestone: none → mitaka-1
Amrith Kumar (amrith) wrote :

Matt, I'm assigning this to you since I notice that the BP on the same topic is assigned to you.

Changed in trove:
assignee: nobody → Matthew Van Dijk (mvandijk)
Changed in trove:
importance: Undecided → Low
milestone: mitaka-1 → mitaka-3
Matthew Van Dijk (mvandijk) wrote :

The fix is in. I no longer see this.

Amrith Kumar (amrith) on 2016-03-11
Changed in trove:
milestone: mitaka-3 → mitaka-rc1
Doug Shelley (0-doug) on 2016-03-11
Changed in trove:
status: Confirmed → Fix Committed
Amrith Kumar (amrith) on 2016-04-05
Changed in trove:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers