Red Hat Bugzilla – Bug 1301673
DHCP-DDNS server UDP socket resource temporarily unavailable
Last modified: 2017-12-23 04:14:57 EST
Description of problem:
After enabling and starting the kea-dhcp-ddns service the following messages are logged repeatedly to /var/log/kea-ddns.log:
2016-01-25 10:08:22.131 ERROR [kea-dhcp-ddns.libdhcp-ddns/1468] DHCP_DDNS_NCR_UDP_RECV_ERROR UDP socket receive error while listening for DNS Update requests: Resource temporarily unavailable
2016-01-25 10:08:22.131 ERROR [kea-dhcp-ddns.dhcp-to-d2/1468] DHCP_DDNS_QUEUE_MGR_RECV_ERROR application's queue manager was notified of a request receive error by its listener.
2016-01-25 10:08:22.131 INFO [kea-dhcp-ddns.dhcpddns/1468] DHCP_DDNS_QUEUE_MGR_RECOVERING application is attempting to recover from a queue manager IO error
Setting the log level to fatal in the config squelches the messages, but the process consumes an entire core. No attempt made to verify if it's actually functioning in this state.
Version-Release number of selected component (if applicable):
Install and run service
Steps to Reproduce:
1. dnf -y install kea
2. systemctl enable kea-dhcp-ddns
3. systemctl start kea-dhcp-ddns
Hundreds of megabytes of errors logged to /var/log/kea-ddns.log in a few minutes.
Successful service start.
The Kea team has determined this is an issue with gcc incorrectly optimizing out some critical logic in the boost::asio function.
Behavior still present in kea-1.2.0-5.fc27.x86_64 package.