| 1 |
features: |
| 2 |
- | |
| 3 |
Mistral engine now supports graceful scale-in. That is, if a number of |
| 4 |
engines in a cluster needs to be reduced manually it is now possible to |
| 5 |
do w/o breaking currently running workflows. In order to shutdown a |
| 6 |
Mistral engine, SIGTERM signal needs to be sent to the corresponding |
| 7 |
process. In Unix operatating systems it's a matter of running the command |
| 8 |
"kill <engine PID>" in any shell. When this signal is caught by the |
| 9 |
process, it has a certain amount of time configured by the |
| 10 |
'graceful_shutdown_timeout' property to complete currently running |
| 11 |
database transactions and process all buffered RPC messages that have |
| 12 |
already been polled from the queue. After this time elapses, the process |
| 13 |
will be forced to exit. |
| 14 |
By default, the value of the 'graceful_shutdown_timeout' property is |
| 15 |
60 (seconds). |