This name is already in the /etc/hosts file. I am attempting to listen on localhost. source local { network(localip(localhost) transport("tcp") port(1514) max_connections(100) log_fetch_limit(500) log_iw_size(10000) flags(no-multi-line,syslog-protocol) tags(uvic.local) keep-timestamp(no) ); }; and the /etc/hosts file has 127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4 and on reboot ... 05:13:28.891 host daemon.notice systemd[1]: syslog-ng.service: main process exited, code=exited, status=2/INVALIDARGUMENT 05:13:28.891 host daemon.err systemd[1]: Failed to start System Logger Daemon. 05:13:28.891 host daemon.notice systemd[1]: Unit syslog-ng.service entered failed state. 05:13:28.891 host daemon.warning systemd[1]: syslog-ng.service failed. 05:13:29.007 host daemon.info systemd[1]: syslog-ng.service holdoff time over, scheduling restart. 05:13:29.009 host daemon.info systemd[1]: Stopped System Logger Daemon. 05:13:29.009 host daemon.info systemd[1]: Starting System Logger Daemon... 05:13:29.190 host daemon.info syslog-ng[6366]: [05:13:29.189991] Error resolving hostname; host='localhost' 05:13:29.197 host daemon.notice systemd[1]: syslog-ng.service: main process exited, code=exited, status=2/INVALIDARGUMENT 05:13:29.197 host daemon.err systemd[1]: Failed to start System Logger Daemon. 05:13:29.197 host daemon.notice systemd[1]: Unit syslog-ng.service entered failed state. 05:13:29.198 host daemon.warning systemd[1]: syslog-ng.service failed. 05:13:29.345 host daemon.info systemd[1]: syslog-ng.service holdoff time over, scheduling restart. 05:13:29.346 host daemon.info systemd[1]: Stopped System Logger Daemon. 05:13:29.346 host daemon.info systemd[1]: Starting System Logger Daemon... 05:13:29.412 host daemon.info syslog-ng[6457]: [05:13:29.412745] Error resolving hostname; host='localhost' 05:13:29.415 host daemon.notice systemd[1]: syslog-ng.service: main process exited, code=exited, status=2/INVALIDARGUMENT 05:13:29.415 host daemon.err systemd[1]: Failed to start System Logger Daemon. 05:13:29.415 host daemon.notice systemd[1]: Unit syslog-ng.service entered failed state. 05:13:29.415 host daemon.warning systemd[1]: syslog-ng.service failed. 05:13:29.556 host daemon.info systemd[1]: syslog-ng.service holdoff time over, scheduling restart. 05:13:29.556 host daemon.info systemd[1]: Stopped System Logger Daemon. 05:13:29.557 host daemon.info systemd[1]: Starting System Logger Daemon... 05:13:29.623 host daemon.info syslog-ng[6487]: [05:13:29.623206] Error resolving hostname; host='localhost' 05:13:29.626 host daemon.notice systemd[1]: syslog-ng.service: main process exited, code=exited, status=2/INVALIDARGUMENT 05:13:29.626 host daemon.err systemd[1]: Failed to start System Logger Daemon. 05:13:29.626 host daemon.notice systemd[1]: Unit syslog-ng.service entered failed state. 05:13:29.626 host daemon.warning systemd[1]: syslog-ng.service failed. 05:13:29.806 host daemon.info systemd[1]: syslog-ng.service holdoff time over, scheduling restart. 05:13:29.807 host daemon.info systemd[1]: Stopped System Logger Daemon. 05:13:29.807 host daemon.info systemd[1]: Starting System Logger Daemon... 05:13:29.867 host daemon.info syslog-ng[6537]: [05:13:29.867292] Error resolving hostname; host='localhost' 05:13:29.870 host daemon.notice systemd[1]: syslog-ng.service: main process exited, code=exited, status=2/INVALIDARGUMENT 05:13:29.870 host daemon.err systemd[1]: Failed to start System Logger Daemon. 05:13:29.870 host daemon.notice systemd[1]: Unit syslog-ng.service entered failed state. 05:13:29.870 host daemon.warning systemd[1]: syslog-ng.service failed. 05:13:29.893 host daemon.info network[6435]: Bringing up loopback interface: [ OK ] 05:13:30.009 host daemon.info systemd[1]: syslog-ng.service holdoff time over, scheduling restart. 05:13:30.009 host daemon.info systemd[1]: Stopped System Logger Daemon. 05:13:30.009 host daemon.warning systemd[1]: start request repeated too quickly for syslog-ng.service 05:13:30.010 host daemon.err systemd[1]: Failed to start System Logger Daemon. 05:13:30.010 host daemon.notice systemd[1]: Unit syslog-ng.service entered failed state. 05:13:30.010 host daemon.warning systemd[1]: syslog-ng.service failed. 05:13:34.214 host daemon.info network[6435]: Bringing up interface eth0: [ OK ] 05:13:34.242 host daemon.info systemd[1]: Reached target Network. 05:13:37.651 host lpr.info cupsd[6665]: Listening to 127.0.0.1:631 on fd 10... After this (when I can log in) a simple "service syslog-ng start" will correctly start the syslog-ng service. the rsyslog package does have systemd unit file definitions of [Unit] Wants=network.target network-online.target After=network.target network-online.target but that might be Redhat's way of saying that systemd-journal is good enough when you don't have networking since [Install] WantedBy=multi-user.target is also present meaning that rsyslog is only started when a multi-user boot is done. Since the syslog-ng.service file also includes [Install] WantedBy=multi-user.target I think that it is reasonable to add the unit definitions of [Unit] Wants=network.target network-online.target After=network.target network-online.target and the problem just goes away for normal boot up environments. I think I'll post this as a packaging systemd unit file definition bug. On 4/17/19 11:29 PM, Scheidler, Balázs wrote:
You mean bind address is specified by dns? You could add that to the hosts file.
Bazsi
On Wed, Apr 17, 2019, 23:48 Evan Rempel <erempel@uvic.ca <mailto:erempel@uvic.ca> wrote:
I've been looking at an interesting scenario on Redhat 7 (systemd) where syslog-ng fails to start because it can not resolve a network source name. This happens because the network has not started. Systemd attempts to restart syslog-ng a few times and then decides that it is restarting syslog-ng too quickly and gives up on it.
2019-04-17T05:13:30.009 daemon.warning systemd[1]: start request repeated too quickly for syslog-ng.service 2019-04-17T05:13:30.010 daemon.err systemd[1]: Failed to start System Logger Daemon. 2019-04-17T05:13:30.010 daemon.notice systemd[1]: Unit syslog-ng.service entered failed state. 2019-04-17T05:13:30.010 daemon.warning systemd[1]: syslog-ng.service failed.
My first reaction was to add the "Wants=network-online.target" to the service definition to make syslog-ng start after the network has started, but that would mean that if I boot up without a network that syslog-ng would never start, and that seems incorrect to me.
Is there a way to to configure a network source to be optional and continue to retry binding to it after startup?
How do others handle this issue?
-- Evan
______________________________________________________________________________ Member info: https://lists.balabit.hu/mailman/listinfo/syslog-ng Documentation: http://www.balabit.com/support/documentation/?product=syslog-ng FAQ: http://www.balabit.com/wiki/syslog-ng-faq
______________________________________________________________________________ Member info: https://lists.balabit.hu/mailman/listinfo/syslog-ng Documentation: http://www.balabit.com/support/documentation/?product=syslog-ng FAQ: http://www.balabit.com/wiki/syslog-ng-faq
-- Evan Rempel erempel@uvic.ca Senior Systems Administrator 250.721.7691 Data Centre Services, University Systems, University of Victoria