As a confirmation, the profile which corrupted was exclusively WSS connections.
On Jan 7, 2018, at 4:24 PM, Jeremy Lainé <jeremy.laine@m4x.org> wrote:
Hi all,
Just as Geoff Mina recently reported, I too have encountered numerous instances of sqlite database corruptions, without any power cuts or disk failures. I have experienced this both on EC2 instances and on physical servers.
I saw António Silva asked whether Geoff Mina was using wss endpoints, could you elaborate on why this might be the root cause? In the setups where I encountered the corruption, I am indeed using wss endpoints. How about you Geoff?
As a workaround I am considering migrating to postgresql (core + sofia) but it's frustating not to understand the root cause of the problem.
Cheers,
Jeremy