adcli 0.9.0 (1b152803) discovery coredumps when host in _ldap._tcp is unresponsive
Hello,
I am running openSUSE Tumbleweed 20210320 with adcli 0.9.0 ( 1b152803 ) and have observed the following behavior. This does not occur with an older release (0.8.1).
ginnungagap:~ # adcli -v info redacted.edu
* Discovering domain controllers: _ldap._tcp.redacted.edu
* Sending NetLogon ping to domain controller: ad104.redacted.edu
* Sending NetLogon ping to domain controller: ad901.redacted.edu
! Couldn't perform discovery search: Can't contact LDAP server
adcli: result.c:111: ldap_result: Assertion `ld != NULL' failed.
Aborted (core dumped)
This affects other operations such as "update" as well.
This occurs when trying to connect to "ad901" consistently, and adcli rolls through every host returned in the SRV record "_ldap._tcp" so it always hits this hostname. The host "ad901" is offline as it is a DR system that is normally kept powered down. If I manually specify any of the other controllers with the "-S" argument, adcli operates correctly.
The normal operation I would expect would be to ignore connection errors until all hosts are tried, and propagate to a failure only if none were reachable. This is what occurred with previous releases (such as 0.8.1, which I have available on this network on a RHEL7 host for comparison).
Please excuse me if this is a resolved / known issue - I did search and was not able to find this specific issue.