After a failed connection attempt to the SMTP server, the wake-up time of the ALM engine gets stuck in the future (31/12/2036) and no mail alerts are sent anymore.
The following error appears in universe.log:
| 2024-05-03 13:13:08 |ERROR|X|IO |pid=1052192.140601055934208| u_send_mail | Cannot connect SMTP server <smtp_server_address> on port 25
| 2024-05-03 13:13:08 |ERROR|X|IO |pid=1052192.140601055934208| NOTIFY_SMTP | [<NODE>] NOTIFICATION_EMAIL [SEND ERROR 7] / Dollar Univer$e: [Incidenté] [<TSK>] [] [<UPROC>] [<MU>] [<UPR_NUM>] [<LAUNCH_NUM>] [] [X]
| 2024-05-03 13:20:16 |WARN |X|IO |pid=1052192.140601718728448| o_lock_engines_during_pur | Lock timeout occurred (10 min.).
| 2024-05-03 13:20:16 |WARN |X|IO |pid=1052192.140601718728448| o_lock_engines_during_pur | Module Alerting still busy.
Due to the stuck ALM, the node doesn't stop with 'unistop' only 'unistop -force' can stop the node.
Defect
Update to a fix version listed below or a newer version if available.
Fix version:
Component(s): Application Server
Dollar Universe 7.01.01 - Planned release Q4 2024
After applying the fix, make sure to set the newly added variable U_SMTP_SOCKET_TIMEOUT with a value like 10 seconds, as else the behavior will remain the same as before.
The reference for this issue is DU_AS-7553
Public Description: Added a timeout on socket reading when discussing with SMTP server U_SMTP_SOCKET_TIMEOUT with value 0 by default.