We are sorry for the inconvenience caused by the incident of 24/10/2019 at 16h.
This problem is due to an overflow of the database's useful memory, which has caused certain processes to stop.
Some processes restarted automatically, others had to be restarted manually.
We had to shut down the front servers before restarting them 1 to 1.
No data was lost. The transactions could continue.
After a period of increased monitoring, the incident was closed at 17:46.
The action plan defined following this incident has two main areas of improvement: