Hide Forgot
Gregg, Is this still a thing that can happen? Is there a direction we can investigate or at least try to figure out how likely this will be so we can move it away from being a blocker?
We've seen this happen in the past when a replication worker is terminated abnormally and some pending changes are lost. Since this bug was reported, we've made some improvements to rubyrep to make it more stable which would mitigate this particular issue. Also, since this is an issue specific to rubyrep and rubyrep replication has been completely removed from the application in 5.7, it shouldn't be a blocker for the errata. I propose closing this BZ. Alex, please close if you are in agreement. Thanks
Agreed.