URL:
https://github.com/SSSD/sssd/pull/5532
Title: #5532: ldap: retry ldap_install_tls() when watchdog interruption
alexey-tikhonov commented:
"""
Hi,
thank you for updates and logs.
There is a weird thing in the log:
```
[sss_ldap_init_sys_connect_done] (0x0020): Assuming TLS handshake was interrupted
[sdap_sys_connect_done] (0x0020): sdap_async_connect_call request failed: [1432158320]:
TLS handshake was interrupted.
[sdap_cli_connect_done] (0x0040): TLS handshake was interruped, provider may retry
[be_resolve_server_send] (0x0040): Will not retry, maximum number of attempts (2)
reached
[fo_resolve_service_send] (0x0100): Trying to resolve service 'LDAP'
[get_server_status] (0x1000): Status of server '10.0.155.220' is
'working'
[get_port_status] (0x1000): Port status of port 636 for server '10.0.155.220' is
'working'
[fo_resolve_service_activate_timeout] (0x2000): Resolve timeout set to 6 seconds
[get_server_status] (0x1000): Status of server '10.0.155.220' is
'working'
[be_resolve_server_process] (0x0040): The fail over cycled through all available servers
[be_resolve_server_done] (0x1000): Server resolution failed: [2]: No such file or
directory
```
-- `status ... is 'working'` despite `Will not retry`
I guess the reason is `be_fo_set_port_status(... PORT_NOT_WORKING)` isn't executed in
this case (while it should)
"""
See the full comment at
https://github.com/SSSD/sssd/pull/5532#issuecomment-810337498