Hi All ;)
I am using a virtual machine with: Red Hat Enterprise Linux Server release 5.7 (Tikanga) Linux logserver01 2.6.18-274.el5 #1 SMP Fri Jul 8 17:36:59 EDT 2011 x86_64 x86_64 x86_64 GNU/Linux syslog-ng 2.1.4
I have following configuration:
... 9 options { 10 create_dirs (yes); 11 dir_group (root); 12 dir_owner (root); 13 dir_perm (0700); 14 group (root); 15 owner (root); 16 perm (0600); 17 flush_lines(1); 18 flush_timeout (1000); 19 keep_hostname (yes); 20 log_fifo_size (1); 21 use_dns (no); 22 use_fqdn (no); 23 }; ... 39 source s_stdout { 40 # file ("/logs/stdout.log" flags(no-parse) follow_freq(1)); 41 file ("/logs/stdout.log" flags(no-parse)); 42 }; ... 61 destination d_stdout { file("/var/log/$YEAR$MONTH$DAY/stdout"); }; ... 80 log { source(s_stdout); destination(d_stdout); };
The problem is that changes in /logs/stdout.log are only visible in /var/log/20140717/stdout when I perform /etc/init.d/syslog-ng reload or restart. I tried several settings of flush_*, log_fifo_size and follow_freq but with no luck :D
Is it a problem withe the version that I use (quite old :D ) or is there a mistake in my configuration maybe?
BR, Rafal.
On 7/17/2014 6:19 AM, Rafał Radecki wrote:
The problem is that changes in /logs/stdout.log are only visible in /var/log/20140717/stdout when I perform /etc/init.d/syslog-ng reload or restart. I tried several settings of flush_*, log_fifo_size and follow_freq but with no luck :D
Is it a problem withe the version that I use (quite old :D ) or is there a mistake in my configuration maybe?
afaik, syslog-ng isn't part of CentOS 5, and there is no top level directory /logs/ at all, logging is done to /var/log and the syslog config files are in /etc, so you must have a heavily customized system configuration. Since noone can possibly know what all you've done, it seems like you're on your own here.
On 2014-07-17, John R Pierce pierce@hogranch.com wrote:
On 7/17/2014 6:19 AM, Rafał Radecki wrote:
The problem is that changes in /logs/stdout.log are only visible in /var/log/20140717/stdout when I perform /etc/init.d/syslog-ng reload or restart. I tried several settings of flush_*, log_fifo_size and follow_freq but with no luck :D
Is it a problem withe the version that I use (quite old :D ) or is there a mistake in my configuration maybe?
afaik, syslog-ng isn't part of CentOS 5, and there is no top level directory /logs/ at all, logging is done to /var/log and the syslog config files are in /etc, so you must have a heavily customized system configuration. Since noone can possibly know what all you've done, it seems like you're on your own here.
syslog-ng is available in EPEL for RHEL 5. Interestingly, it appears to log to /var/lib/syslog-ng. (I don't use syslog-ng; I'm just going by the output of 'repoquery -l syslog-ng'.)
On 7/17/2014 11:53 AM, Liam O'Toole wrote:
syslog-ng is available in EPEL for RHEL 5. Interestingly, it appears to log to /var/lib/syslog-ng. (I don't use syslog-ng; I'm just going by the output of 'repoquery -l syslog-ng'.)
and EPEL is not /part of centos/.
anyways, I suspect the OP's question would best be handled on an syslog-ng mail list