#1933 sssd fails to resolve hosts/services once the network is up
Closed: Fixed None Opened 6 years ago by georgiou.

After boot once the network is up sssd fails to resolve krb5/ldap hosts/services. From the logs it looks like it tries DNS resolves before the network is up but it doesn't at the time it detects the /etc/resolv.conf change. After a few minutes it retries and is able to resolve everything and it brings the domain online.

This is on a fully updated Fedora 17 machine(s) using the network scripts and not NM, the ssd rpm is: sssd-1.9.4-2.fc17.x86_64


This is the interesting part of the logs:

[sbus_message_handler] (0x4000): Received SBUS method [resetOffline]
[check_if_online] (0x2000): Trying to go back online!
[set_srv_data_status] (0x0100): Marking SRV lookup of service 'KERBEROS' as 'neutral'
[set_srv_data_status] (0x0100): Marking SRV lookup of service 'KERBEROS' as 'neutral'
[set_server_common_status] (0x0100): Marking server 'ads-wn-2.foo.com' as 'name not resolved'
[fo_set_port_status] (0x0100): Marking port 0 of server 'ads-wn-2.foo.com' as 'neutral'
[set_server_common_status] (0x0100): Marking server 'ads-wn-1.foo.com' as 'name not resolved'
[fo_set_port_status] (0x0100): Marking port 0 of server 'ads-wn-1.foo.com' as 'neutral'
[set_srv_data_status] (0x0100): Marking SRV lookup of service 'LDAP' as 'neutral'
[fo_resolve_service_send] (0x0100): Trying to resolve service 'LDAP'
[get_port_status] (0x1000): Port status of port 0 for server '(no name)' is 'not working'
[fo_resolve_service_send] (0x0020): No available servers for service 'LDAP'
[be_resolve_server_done] (0x1000): Server resolution failed: 5
[check_online_callback] (0x0100): Backend returned: (1, 0, <NULL>) [Provider is Offline

Even though the back end receives the resetOffline signal and sets the port
status and server status ro "neutral" and "name not resolved" respectively,
the next resolve request shortcuts to "not working again".

We should investigate along with fixing #1886, those might be very similar if not the same.

Fields changed

milestone: NEEDS_TRIAGE => SSSD 1.10.0
priority: major => minor

Fields changed

rhbz: => todo

Fields changed

milestone: SSSD 1.10.0 => SSSD 1.10.1

Fields changed

changelog: =>
owner: somebody => mzidek
patch: 0 => 1

I proposed to fix the problem in the sssd-1-9 branch as well on the sssd-devel list.

Fields changed

resolution: => fixed
status: new => closed

Fields changed

changelog: => N/A, just a bugfix

Metadata Update from @georgiou:
- Issue assigned to mzidek
- Issue set to the milestone: SSSD 1.10.1

2 years ago

Login to comment on this ticket.

Metadata