- syslog-ng - lists.balabit.hu

syslog-ng dumps core (twice yesterday) how do I investigate cause?
by David Lacerte 01 Mar '21

01 Mar '21

Running syslog-ng-3.5.4.1/ on CentOS 6.5 started seeing core dumps in var.log/kern approximately once per week. The "supervising syslog-ng" process seems to restart everything so the problem is not really being noticed (except we have gaps, pauses during which logging for some systems is not occurring) Here is /var/log/kern from yesterday showing two process core dumps: Feb 28 18:05:03 sl.its.uwo.ca kernel: syslog-ng[45160]: segfault at 0 ip 00007f0a767d0acc sp 00007f0a6fffca50 error 6 in libsyslog-ng-3.5.4.1.so[7f0a7679e000+8b000] Feb 28 23:04:02 sl.its.uwo.ca kernel: syslog-ng[5442]: segfault at 0 ip 00007f0a767d0acc sp 00007f0a6fffca50 error 6 in libsyslog-ng-3.5.4.1.so[7f0a7679e000+8b000] What is the approach I should use to investigate this (seemingly random) seg fault problem? Dave Lacerte UWO.ca

2 1

Console logging stops after some time.
by Alexandre Santos 24 Feb '21

24 Feb '21

Hi All, I am using a Debian10 with syslog-ng 3.19. I have a simple configuration formed by a log to a file and a log to the console "usertty(*)". Please check the attached file. *After some time the logging to the console stops, but the logging to the file keeps working.* I used strace to debug this. (strace -p <syslog-ng-pid> -yy -f -s 256 -e trace=write) *root@localhost:~# w root 12:50:41 up 3:16, 5 users, load average: 0.21, 0.35, 0.37USER TTY FROM LOGIN@ IDLE JCPU PCPU WHATroot ttyS0 - 09:36 45:18 0.02s 0.02s -bashroot pts/1 10.0.2.2 09:40 1.00s 0.26s 0.00s w rootroot pts/4 10.0.2.2 09:56 24:08 3.03s 2.84s strace -p 23429 -yy -f -s 256 -e trace=writeroot pts/5 10.0.2.2 10:26 22:01 0.01s 0.01s tail -f /var/log/apps.logroot pts/0 10.0.2.2 12:31 19:25 21.77s 0.00s sleep 60root@localhost:~# strace -p 23429 -yy -f -s 256 -e trace=write[pid 24379] write(26</dev/ttyS0<char 4:64>>, "2021 Feb 19 12:31:16 localhost 8389:Usm INFO {Orchestrator.cpp:1329} attr_id=10931 moid=8398\n", 107) = 107[pid 24379] write(26</dev/pts/1<char 136:1>>, "2021 Feb 19 12:31:16 localhost 8389:Usm INFO {Orchestrator.cpp:1329} attr_id=10931 moid=8398\n", 107) = 107[pid 24379] write(26</dev/pts/4<char 136:4>>, "2021 Feb 19 12:31:16 localhost 8389:Usm INFO {Orchestrator.cpp:1329} attr_id=10931 moid=8398\n", 1072021 Feb 19 12:31:16 localhost 8389:Usm INFO {Orchestrator.cpp:1329} attr_id=10931 moid=8398[pid 24379] write(26</dev/pts/4<char 136:4>>, "2021 Feb 19 12:31:16 localhost 8389:Usm INFO {Orchestrator.cpp:1329} attr_id=10931 moid=8398\n", 1072021 Feb 19 12:31:16 localhost 8389:Usm INFO {Orchestrator.cpp:1329} attr_id=10931 moid=8398[pid 24379] write(26</dev/pts/5<char 136:5>>, "2021 Feb 19 12:31:16 localhost 8389:Usm INFO {Orchestrator.cpp:1329} attr_id=10931 moid=8398\n", 107) = 107[pid 24379] write(26</dev/pts/0<char 136:0>>, "2021 Feb 19 12:31:16 localhost 8389:Usm INFO {Orchestrator.cpp:1329} attr_id=10931 moid=8398\n", 107) = 107[pid 24379] write(23<pipe:[1957563]>, "<190>1 2021-02-19T12:31:16.855+00:00 localhost usm 6802 - - 8389:Usm INFO {Orchestrator.cpp:1329} attr_id=10931 moid=8398\n", 136) = 136[pid 24379] write(23<pipe:[1957563]>, "<190>1 2021-02-19T12:31:16.885+00:00 localhost usm 6802 - - 8395:Usm INFO {Orchestrator.cpp:1329} attr_id=10931 moid=8398\n", 136) = 136[pid 23703] write(1</var/log/apps.log>, "<190>1 2021-02-19T12:31:16.855+00:00 localhost usm 6802 - - 8389:Usm INFO {Orchestrator.cpp:1329} attr_id=10931 moid=8398\n", 136) = 136[pid 23703] write(1</var/log/apps.log>, "<190>1 2021-02-19T12:31:16.885+00:00 localhost usm 6802 - - 8395:Usm INFO {Orchestrator.cpp:1329} attr_id=10931 moid=8398\n", 136) = 136* After some time I stop seeing writes to the console's, and only see writes to the apps.log file. *[pid 24379] write(23<pipe:[1957563]>, "<190>1 2021-02-19T12:31:21.334+00:00 localhost dbf 7884 - - 8159:Dbf INFO {transaction_context.cpp:20} dtor of TransCtxt, tid: 120\n", 145) = 145[pid 23703] write(1</var/log/apps.log>, "<190>1 2021-02-19T12:31:21.334+00:00 localhost dbf 7884 - - 8159:Dbf INFO {transaction_context.cpp:20} dtor of TransCtxt, tid: 120\n", 145) = 145* It seems that due to the high amount of messages the logging to the console is discarded. Can you help me to solve this, Thanks in advance, Alex

2 5

query on using throttle in syslog-ng.conf file
by SIMON BABY 20 Feb '21

20 Feb '21

Hello Team, I am new to this group and I have a query on adding the throttle attribute in syslog-ng.conf file. My target is little slow to process all the messages sent by the sender and sometimes the link connected to the target device is broken. I am thinking of slowing down the sender by adding the throttle attribute. I have the below queries: 1) What exactly the throttle confoguration does? 2) What does throttle(500) mean ? will it send 500 Bytes per second or 500 messages per second? What does the message here mean ? Can it be the entire message sent by the application ? Is there an upper limit and lower limit ? 3) Any side effect of my system if I am going to use throttle(). 4) Any other method in syslog-ng to delay the logs sending at the sender? 5) My destination configuration is below. Is it a valid configuration ? destination logFiler { file("/var/log/wq.log" template("${FULLDATE}${TZ} ${HOST} ${PROGRAM} [$LEVEL] ${MSG}\n") template_escape(yes) throttle(500));}; Thank you for your time. Regards Simon

2 9

tty10 on Debian under LXC
by John Kristoff 16 Feb '21

16 Feb '21

I ran into the exact problem succinctly and completely written up here: <https://www.claudiokuenzler.com/blog/988/syslog-ng-lxc-container-no-space-l…> It doesn't appear this was brought to the attention of the devs nor an issue even filed in github. I'd like to make sure I didn't miss something, so posting here for comment before I submit a github issue. In summary, LXC sets up a small number of tty devices by default. However, syslog-ng in Debian has this config statement: destination d_console_all { file(`tty10`); }; This results in syslog-ng writing the a new /dev/tty10 file instead of to a device, which will pretty quickly fill up an LXC /dev partition, causing syslog-ng to soon start failing when it sees the device associated with tty10 file handle full. I have modified my syslog-ng.conf directly to use tty2 as the blog post suggested, but it would be nice if there was a "cleaner" work-around in the future. John

2 2

Insider 2021-02: proxy protocol; sudo JSON; Kafka;
by Peter Czanik (pczanik) 11 Feb '21

11 Feb '21

Dear syslog-ng users, This is the 88th issue of syslog-ng Insider, a monthly newsletter that brings you syslog-ng-related news. NEWS Finding the real source IP: using the PROXY protocol ---------------------------------------------------- Until now collecting logs behind proxies or load balancers needed some compromises. You either trusted the host information included in the log messages or you could only see the proxy as the sender host. Starting with syslog-ng 3.30 there is a third option available: using the PROXY protocol. While not an official Internet standard, it is supported by a number of popular software, like HAProxy. Other software can be extended to use it, like F5 load balancers using iRules. This way crucial information about the original network connection is not lost, but it is forwarded to the server by the proxy. https://www.syslog-ng.com/community/b/blog/posts/finding-the-real-source-ip… Parsing sudo JSON logs: building a syslog-ng configuration ---------------------------------------------------------- The latest version of sudo, version 1.9.4 includes support for JSON formatted logging. Compared to traditional sudo logs, it has the advantage of containing more information in a structured way. While traditional sudo logs are also parsed automatically by syslog-ng, it is worth taking a look at the new JSON formatted logs. >From this blog, you can learn how the new logs look like and also a configuration working with these logs. Instead of just posting a complex configuration, I try to show you how my configuration was built. Creating a new configuration in smaller iterations makes the resulting configurations easier to debug. https://www.syslog-ng.com/community/b/blog/posts/parsing-sudo-json-logs-bui… Kafka destination improved with template support ------------------------------------------------ The C implementation of the Kafka destination in syslog-ng has been improved in version 3.30. Support for templates in topic names was added as a result of a Google Summer of Code (GSoC) project. The advantage of the new template support feature is that you no longer have to use a static topic name. For example, you can include the name of your host or the application sending the log in the topic name. From this blog you can learn about a minimal Kafka setup, configuring syslog-ng and testing syslog-ng with Kafka. https://www.syslog-ng.com/community/b/blog/posts/kafka-destination-improved… Syslog-ng PE 7.0.23 released ---------------------------- Version 7.0.23 of syslog-ng PE was released with clustering support for Windows Event Collector (WEC). You can learn more about it from the documentation or from this short video: https://www.youtube.com/watch?v=bYdIJaM24Z8 WEBINARS * You can browse recordings of past webinars at https://www.syslog-ng.com/events/ Your feedback and news, or tips about the next issue are welcome. To read this newsletter online, visit: https://syslog-ng.com/blog/ Peter Czanik (CzP) <peter.czanik(a)oneidentity.com> Balabit (a OneIdentity company) / syslog-ng upstream https://syslog-ng.com/community/ https://twitter.com/PCzanik

1 0

flags(final) - Question
by Matthias Gruber 09 Feb '21

09 Feb '21

Hi! After searching why a log-message doesnt got the right destination path, I propably got a bit crossed up with the flag(final) statement, perhaps someone can enlight me :-) Following scenario (no valid syntax, only for description) filter(f_a) { match("something");}; destination(d_a) { file("/destination_a"); }: destination(d_b) { file("/destination_b"); }: destination(d_c) { file("/destination_c"); }: log { source(s_remote_default_syslog_udp); if filter(f_a) { destination(d_a); flag(final); } else { destination(d_b); flag(final); }; destination(d_c); }; What I expect is if filter->f_a hits the message is written in destination->d_a (what he do), then hits the final, and ends the log-statement What I discovered he also write destination->d_c Is that the expected behaviour? And if so, how to accomplish, that e.g. if the if or the else part hit, no destination -> d_c is written. Yeah I know, why the destination->d_c :-), correct I stumbled over this, normally after such a if/elif/else-Block I dunno have this destination->d_c line in my syslog-ng environments. Only a bit courious about it and seeking for an explanation. cheers Matthias ------------------------------------------------------------------------------------ METZLER Informationstechnologie Matthias Gruber IT-Infrastruktur & -Betrieb B. Metzler seel. Sohn & Co. Kommanditgesellschaft auf Aktien Untermainanlage 1 60329 Frankfurt am Main Telefon (0 69) 21 04 - 43 30 Telefax (0 69) 21 04 - 40 40 MGruber(a)metzler.com www.metzler.com Persönlich haftende Gesellschafter: Harald Illy, Friedrich von Metzler, Emmerich Müller, Gerhard Wiesheu Vorsitzender des Aufsichtsrats: Dr. Christoph Schücking Sitz der Gesellschaft: Frankfurt am Main, Handelsregister-Nr. HRB 27 515 Diese Nachricht ist vertraulich. Sollten Sie nicht der vorgesehene Empfänger sein, so bitten wir Sie höflich, dies unverzüglich dem Absender mitzuteilen und die Nachricht zu löschen. Es ist unzulässig, die Nachricht unbefugt weiterzuleiten oder zu kopieren. Da wir nicht die Echtheit oder Vollständigkeit der in dieser Nachricht enthaltenen Informationen garantieren oder zusichern können, sind die vorstehenden Ausführungen rechtlich nicht bindend. Eine Haftung hierfür wird ausgeschlossen. This message is confidential. If you are not the intended recipient, we kindly ask you to inform the sender and delete the information. Any unauthorised dissemination or copying hereof is prohibited. As we cannot guarantee or assure the genuineness or completeness of the information contained in this message, the statements set forth above are not legally binding. Accordingly we cannot accept any liability for their contents.

2 2

filtering based on text?
by Dan Egli 08 Feb '21

08 Feb '21

Hey folks, I've looked for a way to do this, and I guess my google-foo is weak today. I've got a situation where on one of my machines, an automated process FREQUENTLY calls sudo so it can gain the permissions it needs to do certain tasks. But that means my /var/log/messages is getting FILLED with sudo messages. I was hoping I could insert some kind of text filter that would allow me to shunt messages where one user (the automated process) calls sudo into another log file. It would basically need to be a nested filter, i.e.: If message_source = sudo then if user = X then log to auto_sudo.log If someone knows how this can be done, I'd appreciate it. Thanks! --- Dan

3 3

PostgreSQL destination - connection refused
by victorsugo 06 Feb '21

06 Feb '21

Hi, I'd like to store incoming log messages in a postgres db. Both db server and syslog-ng run in a docker container. I'm using docker-compose to manage containers. The syslog-ng is depending on the pg container, but... every time I start containers with docker-compose up -d, the syslog-ng starts, but can't connect to database. I found this in the /var/log/messages on the syslog-ng's container: err syslog-ng[1]: Error establishing SQL connection; type='pgsql', host='pgsrv.infra-net', port='', username='syslog', database='postgres', error='could not connect to server: Connection refused ... Is there any way to reconnect automatically or abort syslog-ng if it can't connect to the db? It is important if I don't want to lose some logs... (or do I need to forget the SQL backend if I want to operate a log server without loss of data??) Regards, Sent with [ProtonMail](https://protonmail.com) Secure Email.

1 1

problems with multiline log messages on ubuntu18
by Jerry Scharf 06 Feb '21

06 Feb '21

Hi, We have been struggling with problems with multiple log messages being broken apart. Our apps can send random things to the logger including entire stack traces. We have a classifier that sends unknown lines to their own log file, so this makes a complete mess. This was with the distro apt package. So I went to building 3.25.1 by hand. If I build it with nothing on the configure line, it processes the multiline log messages just fine but if I build it with the ubuntu18 debian rules, it breaks the same way. I narrowed it down to if --eneable-systemd is set it breaks. Here are some relevant things. DISTRIB_DESCRIPTION="Ubuntu 18.04.5 LTS" config cmd ../configure --prefix=/usr/local --sysconfdir=/etc/syslog-ng --enable-dynamic-linking --enable-json --enable-manpages --enable-python --enable-java=no --enable-mongodb=no --enable-legacy-mongodb-options=no SOURCE_REVISION=3.25.1-ir1 --enable-systemd=no attaching the conf and the file classifier (60...) thanks, jerry

2 1

Help with filtering at destination based on custom macro value match
by Akshay Joshi 30 Jan '21

30 Jan '21

My client is sending logs and it has the following config : *template t_global {template("<${PRI}>${LOGHOST} ${MSG_TAG}${MSGHDR}${MSG}\n"); };* *# Global logging remote destination:#-----------------------------------destination d_global_remote { tcp("proxy.dc.nuagedemo.net <http://proxy.dc.nuagedemo.net>" port(10514) template(t_global) tls(peer-verify(required-untrusted) ca-dir('/etc/default/bootstrap/keys') cert_file('/etc/default/bootstrap/keys/cert.pem') key-file('/etc/default/bootstrap/keys/key.pem') ) flags("threaded") );};source s_nuageDiag { file("/home/user/nuage/nuage_diagnostics_daemon.log" follow-freq(10) default-facility(local1) default-priority(info) tags("nuageDiag"));};rewrite w_nuageDiag { set("nuage-diag: ", value("MSG_TAG") condition(tags("nuageDiag"))); };* On the destination, I have this : *source s_network { tcp( port(10514) max-connections(1000) tls( peer-verify(required-untrusted) key-file("/opt/proxy/config/keys/proxy-Key.pem") cert_file("/opt/proxy/config/keys/proxyCert.pem") ca-dir("/opt/proxy/config/keys/proxy-CA.pem") ) );};* *filter nsg_diag { match("nuage-diag: " value("MSG_TAG"));};filter f_messages { (level(info..warn) and filter (nsg_diag)); };* *destination d_logs { file( "/var/log/syslog-ng/logs.txt" owner("root") group("root") perm(0777) );};log { source(s_sys); source(s_network); filter(f_messages); destination(d_logs); };* I can write logs locally without the filtering. But with filtering, it does not match "nuage-diag: " macro. This "MSG_TAG" does not seem to be a standard header but a custom one. I couldn't find many straightforward examples on forums etc.. as well. I am missing a trick or two config-wise for sure. Any pointers / help will be much appreciated. -- Regards, Akshay Joshi

3 2