View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0000768 | Pgpool-II | Bug | public | 2022-09-14 02:31 | 2022-09-29 16:43 |
| Reporter | alexpanayi | Assigned To | pengbo | ||
| Priority | normal | Severity | minor | Reproducibility | always |
| Status | feedback | Resolution | open | ||
| Product Version | 4.3.3 | ||||
| Summary | 0000768: Pgpool auto_failback not working | ||||
| Description | Hey everyone. We are currently running Pgpool 4.3.1 and Postgres 13. We are testing failover and saw there is a setting called: auto_failback = on. From reading the docs, if this is turned on, then Pgpool should be able to re-attach a detached node. However this doesn't seem to be working. We tested several times, to restart a replica. When the replica goes down, the node status is down which is correct. When the replica goes up, it doesn't seem Pgpool is querying the detached node and re-adding it into the pool. Is there another way to automatically re-attach a detatched node into the pool? | ||||
| Tags | No tags attached. | ||||
|
|
Sorry for the late response. To use auto_failback, you need to ensure: - streaming replication check is enabled (https://www.pgpool.net/docs/latest/en/html/runtime-streaming-replication-check.html) - the restarted standby server can connect to primary and streaming replication between primary and standby works properly (connect to primary and check pg_stat_replication) [auto_failback] https://www.pgpool.net/docs/latest/en/html/runtime-config-failover.html#GUC-AUTO-FAILBACK |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2022-09-14 02:31 | alexpanayi | New Issue | |
| 2022-09-29 16:36 | pengbo | Assigned To | => pengbo |
| 2022-09-29 16:36 | pengbo | Status | new => assigned |
| 2022-09-29 16:42 | pengbo | Note Added: 0004124 | |
| 2022-09-29 16:42 | pengbo | Note Edited: 0004124 | |
| 2022-09-29 16:43 | pengbo | Status | assigned => feedback |
