3 pods postgres replication issue after master's fail

[amegz]

I have setuped 3 pod postgres in test cluster.
when i kill a pod with a matser, one of the replicas becomes the new master, kubernetes creates a new pod and it connects to the master, replication is restored.
However, the third pod connects to the new master and can't find the wal files there:

2024-03-27 08:41:40.004 GMT [3473] FATAL: could not receive data from WAL stream: ERROR: requested WAL segment 000000040000000000000013 has already been removed
2024-03-27 08:41:40.004 GMT [28] LOG: waiting for WAL to become available at 0/13000630
2024-03-27 08:41:45.006 GMT [3482] LOG: started streaming WAL from primary at 0/13000000 on timeline 4
2024-03-27 08:41:45.006 GMT [3482] FATAL: could not receive data from WAL stream: ERROR: requested WAL segment 000000040000000000000013 has already been removed
2024-03-27 08:41:45.007 GMT [28] LOG: waiting for WAL to become available at 0/13000630
2024-03-27 08:41:50.008 GMT [3483] LOG: started streaming WAL from primary at 0/13000000 on timeline 4
2024-03-27 08:41:50.008 GMT [3483] FATAL: could not receive data from WAL stream: ERROR: requested WAL segment 000000040000000000000013 has already been removed
2024-03-27 08:41:50.009 GMT [28] LOG: waiting for WAL to become available at 0/13000630

I'm not sure if this is a bug or not, what is the best way to change this behavior so that after the master crashes the replica is restored to its previous state?