"Fix issue of etcd servers running out of space by enabling auto-compaction."
"Release version 2.0.0-alpha.107 and update docs"
"Worker exhaustion (#443)\n\n* Fix for cycling error messages when source disappears.\r\n\r\n* Fix docker settings and include prometheus meter endpoint.\r\n\r\n* Fix a disconnection condition problem.\r\n\r\n* Better handle missing route to destination error, report and eat exceptions at the rail worker level to avoid exhausting workers.\r\n\r\nCo-authored-by: Brett Morien email@example.com"
"Release version 2.0.0-alpha.103 and update docs"
"Additional message sending logs (#442)\n\n* Added a few debug logs around message sending.\r\n\r\n* Additional debug messages.\r\n\r\n* More info in the logs\r\n\r\n* Use safe cast operator in message address cast.\r\n\r\n* Make message logging destination variable more clear.\r\n\r\nCo-authored-by: Brett Morien firstname.lastname@example.org"
"Reintroduce versions 99 and 102 to helm chart index."
"run ticker indepently. fix ordering of release (#441)"
"Release version 2.0.0-alpha.102 and update docs"
"Add timers around addressable lease renewal and abandoning in server."
"Clean up debug log spam."
"Release version 2.0.0-alpha.101 and update docs"
"Better instrumenting (#440)\n\n* Switch etcd Node Directory to use leases instead of a polling approach to expiration.\r\n\r\n* Log messages for connection and disconnection of clients to server, debug information about current connections.\r\n\r\nCo-authored-by: Brett Morien email@example.com"
"Release version 2.0.0-alpha.98 and update docs"