Hi, With a quick test I can verify that there is a new "error" tag in the error message: [2015-12-02T14:28:42.828127] Initializing AMQP destination; vhost='/almafa', host='172.17.0.3', port='5672', exchange='applogs', exchange_type='header' [New Thread 0x7ffff7fdd700 (LWP 40)] [2015-12-02T14:28:42.831050] Worker thread started; driver='d_signupws_amqp#0' [2015-12-02T14:28:42.831158] Running application hooks; hook='1' [2015-12-02T14:28:42.831174] Running application hooks; hook='3' [2015-12-02T14:28:42.831213] syslog-ng starting up; version='3.8.0alpha0' [New Thread 0x7ffff315f700 (LWP 41)] [2015-12-02T14:28:43.833637] Incoming log entry; line='ads asda sda sdas dsa dsa' [2015-12-02T14:28:45.833349] Error during AMQP login; driver='d_signupws_amqp#0', error='a socket error occurred', time_reopen='60' [2015-12-02T14:28:48.837345] Error during AMQP login; driver='d_signupws_amqp#0', error='a socket error occurred', time_reopen='60' Br, Micek On Mon, Nov 30, 2015 at 5:19 PM, Scheidler, Balázs < balazs.scheidler@balabit.com> wrote:
They should be resent. On Nov 30, 2015 4:50 PM, "Al Itchon" <AItchon@boardreader.com> wrote:
So are messages lost when this happens, or are they resent when after reconnection?
*From:* syslog-ng-bounces@lists.balabit.hu [mailto: syslog-ng-bounces@lists.balabit.hu] *On Behalf Of *Scheidler, Balázs *Sent:* Thursday, November 26, 2015 2:47 AM *To:* Syslog-ng users' and developers' mailing list < syslog-ng@lists.balabit.hu> *Subject:* Re: [syslog-ng] rabbitmq errors
Nothing intended limits syslog-ng here. That error message happens when rabbitmq returns an error condition. It definitely should include a more detailed error cause.
Syslog-ng suspends this destination for a few seconds (controlled by time-reopen ()) And then reconnects to the amqp server.
I’m seeing an error in my log indicating that messages aren’t being sent to a queue in RabbitMQ. This happens a few times a week. Here’s the error:
Nov 24 13:29:28 syslog09 syslog-ng[7883]: Network error while inserting into AMQP server; time_reopen='10'
I have about 1000 servers sending tcp messages through a call to /dev/tcp to port on the syslog-ng server. It seems messages aren’t sent to the queue for about 1 or 2 minutes, and then it resumes sending again without error. Could this be a network issue or something within syslog-ng that’s limiting the messages?
______________________________________________________________________________ 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
______________________________________________________________________________ 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