
database connections were made on service initialisation and not on process start. this led to all processes of a service sharing the same connection. in this patch, we create unique dispatch_manager to handle a different engine for each process. we keep default to a single worker as schema needs to be updated to avoid possible race/deadlock conditions Closes-Bug: #1298073 Change-Id: I2ad05e2085c0c0f78653c6354d301d18b8dee121
ceilometer
See the ReleaseNotes document and the project home for more info.
Description
Languages
Python
98.6%
Shell
1.3%
Mako
0.1%