XLOG generation peak
All times in UTC. - 5:39pm: PagerDuty alert: "Postgres is generating XLOG too fast, expect this to cause replication lag" https://gitlab.pagerduty.com/incidents/PGXBWQH - 5:44pm: Peak corroborated. Already starting to trend down ![Captura_de_pantalla_2018-01-21_a_la_s__14.43.50](/uploads/78b2f44d5ea16cc30f5063d780c2874f/Captura_de_pantalla_2018-01-21_a_la_s__14.43.50.png) Some other anormalities at the time: - Postgres IO Utilization ![Captura_de_pantalla_2018-01-21_a_la_s__14.49.23](/uploads/36311bc2cf7c012d5921a88352cd5406/Captura_de_pantalla_2018-01-21_a_la_s__14.49.23.png) - NFS timeout peak for nfs-file-12 ![Captura_de_pantalla_2018-01-21_a_la_s__14.48.53](/uploads/e813e29e0deb5090ce3d64f3985d7093/Captura_de_pantalla_2018-01-21_a_la_s__14.48.53.png) - 5:49pm: Alert resolved. XLOG generation levels back to normal: ![Captura_de_pantalla_2018-01-21_a_la_s__14.56.41](/uploads/d201b9d3dbb91d819d56e09e7ac39b8c/Captura_de_pantalla_2018-01-21_a_la_s__14.56.41.png) /cc @gl-infra
issue