I saw these errors in the alert logs after firewall was implemented by the network team. MRP process terminated therefore primary database was not in sync with the standby database.
Strangely other physical standby databases on the same server were not having these errors.
rfs (PID:2287255): CORRUPTION DETECTED: In redo blocks starting at block 790720 count 2048 for T-2.S-226
2022-10-26T12:55:37.988791+00:00
rfs (PID:2287255): Possible network disconnect with primary database
2022-10-26T12:55:53.725218+00:00
rfs (PID:2287301): krsr_rfs_atc: Identified database type as 'PHYSICAL STANDBY': Client is Foreground (PID:233047)
Deleted Oracle managed file +FRA/STBYDB/ARCHIVELOG/2022_10_26/thread_0_seq_0.1514.1151240153
2022-10-26T12:56:00.898623+00:00
rfs (PID:2287247): CORRUPTION DETECTED: In redo blocks starting at block 329730 count 2048 for T-2.S-271
Deleted Oracle managed file +FRA/STBYDB/ARCHIVELOG/2022_10_26/thread_2_seq_271.802.1151240135
2022-10-26T12:56:02.030527+00:00
Errors in file /u01/app/oracle/diag/rdbms/stbydb/STBYDB1/trace/STBYDB1_rfs_2287247.trc (incident=1066689):
ORA-03137: malformed TTC packet from client rejected: [12569] [124] [] [] [] [] [] []
ORA-00272: error writing archive log +FRA/STBYDB/ARCHIVELOG/2022_10_26/thread_2_seq_271.802.1151240135
ORA-00368: checksum error in redo log block
Incident details in: /u01/app/oracle/diag/rdbms/stbydb/STBYDB1/incident/incdir_1066689/STBYDB1_rfs_2287247_i1066689.rcSOLUTION
The suggestion is to check the status of the following features on the firewall and disable them.
– SQLNet fixup protocol
– Deep Packet Inspection (DPI)
– SQLNet packet inspection
– SQL Fixup
– SQL ALG (Juniper firewall)