Closed Bug 1187295 Opened 9 years ago Closed 9 years ago

aus4 read and write webheads out of sync; blocking release work

Categories

(Infrastructure & Operations :: MOC: Problems, task)

task
Not set
blocker

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: bhearsum, Assigned: rwatson)

Details

The aus4 read and write webheads are out of sync, which means any changes we make on the admin host are not reflected on the public facing notes. Effectively, we cannot make any changes to updates. This is blocking the ongoing Beta as well as nightlies from being published. I suspect this is fallout from bug 1175652.


From the admin host (uses aus4-rw-vip.db.phx1.mozilla.com)
mysql> select rule_id, mapping, channel, product from rules where rule_id=45;
+---------+-----------------------+--------------+---------+
| rule_id | mapping               | channel      | product |
+---------+-----------------------+--------------+---------+
|      45 | Firefox-40.0b7-build1 | beta-cdntest | Firefox |
+---------+-----------------------+--------------+---------+
1 row in set (0.00 sec)


And on a public webhead (uses db-aus-prod-ro, which appears to be 10.8.70.21):
mysql> select rule_id, mapping, channel, product from rules where rule_id=45;
+---------+-----------------------+--------------+---------+
| rule_id | mapping               | channel      | product |
+---------+-----------------------+--------------+---------+
|      45 | Firefox-40.0b6-build1 | beta-cdntest | Firefox |
+---------+-----------------------+--------------+---------+
Probably also worth noting that the admin webhead is ausadm.private.phx1.mozilla.com and the public facing nodes are aus[1-6].webapp.phx1.mozilla.com.
Assignee: nobody → rwatson
The load balancers have been set so both reads and writes go to the master. I will resync the slave while it's offline. You should be unblocked now.
All looks good to me now, thanks folks!
Status: NEW → RESOLVED
Closed: 9 years ago
Resolution: --- → FIXED
What was the cause for the read slaves to be out of sync?
Flags: needinfo?(scabral)
improper backup restoration technique. We have fixed that issue.
Flags: needinfo?(scabral)
Component: MOC: Incidents → MOC: Problems
You need to log in before you can comment on or make changes to this bug.