Re: [rsyslog] Safe to upgrade to 8.6.0

2014-12-10 Thread James Lay
On 2014-12-10 04:33 PM, David Lang wrote: The only problems with 8.6 that we know of are related to running the tests. David Lang On Wed, 10 Dec 2014, James Lay wrote: Hey all, After seeing all the emails back and forth over 8.6, I thought I'd better ask here. I'm currently on 8.4.2: rs

Re: [rsyslog] Safe to upgrade to 8.6.0

2014-12-10 Thread David Lang
The only problems with 8.6 that we know of are related to running the tests. David Lang On Wed, 10 Dec 2014, James Lay wrote: Hey all, After seeing all the emails back and forth over 8.6, I thought I'd better ask here. I'm currently on 8.4.2: rsyslog [8.4.2.ad1-0adiscon1precise1] via apt

[rsyslog] Safe to upgrade to 8.6.0

2014-12-10 Thread James Lay
Hey all, After seeing all the emails back and forth over 8.6, I thought I'd better ask here. I'm currently on 8.4.2: rsyslog [8.4.2.ad1-0adiscon1precise1] via apt I would be upgrading to: (8.6.0.r1-0adiscon2precise1 RSyslog V8-Stable:12.04/precise [amd64]) I have nothing special in my rsys

Re: [rsyslog] How to split syslog from syslog server into separate files?

2014-12-10 Thread Rainer Gerhards
2014-12-10 19:00 GMT+01:00 Walzer, Jeff R : > Quick setup: Have five log sources that send syslog to a syslog server > (running rsyslog), which then sends syslog to Splunk server (also running > rsyslog) > > Trying to split syslog coming from syslog server to Splunk into separate > files. Here's t

[rsyslog] How to split syslog from syslog server into separate files?

2014-12-10 Thread Walzer, Jeff R
Quick setup: Have five log sources that send syslog to a syslog server (running rsyslog), which then sends syslog to Splunk server (also running rsyslog) Trying to split syslog coming from syslog server to Splunk into separate files. Here's the rsyslog on Splunk server: # send all messages from

Re: [rsyslog] mmnormalize improvements

2014-12-10 Thread Brian Knox
Rainer - do you have a link to a quick summary of the changes? Would love to know about them and haven't been paying attention due to other work priorities. Thanks! Brian On Wed, Dec 10, 2014 at 10:01 AM, Rainer Gerhards wrote: > Hi all, > > I have now merged Janmejay's awesome mmnormalize imp

Re: [rsyslog] Stuck omfwd connections

2014-12-10 Thread Tim Smith
As I was typing out the email, it occurred to me that the issue is OS related: Looking at a sending server, A, I saw these messages in dmesg: TCP: Peer 10.2.1.2:514/47081 unexpectedly shrunk window 861404336:861405796 (repaired) The local TCP port, 47081 is the same one that is part of the stuck

[rsyslog] Stuck omfwd connections

2014-12-10 Thread Tim Smith
Hi, I have a pair of Linux/RHEL servers (RHEL 6.x), A and B, that forward logs to multiple destinations: - one copy to Splunk syslog listener - one copy to local flume process over TCP - one copy to a remote RSyslog receiver, X and Y (RHEL 6.x) Forwarding copies to Splunk and Flume works fine. Ho

[rsyslog] mmnormalize improvements

2014-12-10 Thread Rainer Gerhards
Hi all, I have now merged Janmejay's awesome mmnormalize improvements. It is currently available in the master-mmnormalize branch. It would be great if some folks could try out that branch and provide feedack. Note that in order to compile, the git master version of liblognorm is required. Right

Re: [rsyslog] InputTCPMaxSessions and MaxOpenFiles relation

2014-12-10 Thread Rainer Gerhards
2014-12-10 10:11 GMT+01:00 Peter Viskup : > Agree it's not simple, but once providing the possibility to define these > type of limits the developer should take care of some basic set of checks > at least. > I agree we should try to find at least the easy config errors. I'll open a github feature

Re: [rsyslog] InputTCPMaxSessions and MaxOpenFiles relation

2014-12-10 Thread Peter Viskup
Agree it's not simple, but once providing the possibility to define these type of limits the developer should take care of some basic set of checks at least. Would like to propose to project owners to take care of making some notices in documentation till/if the checks will not be implemented. Some