September 2011 - syslog-ng - lists.balabit.hu

Output to file in /proc
by Valentijn Sessink 13 Dec '13

13 Dec '13

Hello list, As most of the functionality of "sshguard", "fail2ban" and others is in syslog-ng, I'm trying to use syslog-ng to add entries to my Linux kernel based firewall (iptables). I'm almost ready to implement one, but I have a question about undeliverable log lines: what happens if a certain file() destination blocks? Will syslog-ng retry to deliver the same message, or will it lose the message? What I'm doing is using the "ipt_recent" module; with this module, one can check if a certain IP address was seen before. For example, the following rule will block you if you were seen 5 or more times ("hitcount 5") in the "violators" list last minute ("seconds 60): iptables -A INPUT -m recent \ --rcheck --hitcount 5 --seconds 60 --name violators -j DROP The fun part is, that you can add IP adresses by stuffing them into a file in /proc: echo '+127.0.0.1' > /proc/net/xt_recent/violators ... will add 127.0.0.1 to the "violators" with the current time. However: echo 'whatever' > /proc/net/xt_recent/violators ... gives an error: Invalid argument (22) As far as I can see, syslog-ng will not try again to deliver the same message; but is this by design? I.e. can I trust syslog-ng to not "block" because of a single malformed IP address? Because then, an implementation of sshguard could be simply swapped with a simple: destination df_ipt_recent { file ("/proc/net/xt_recent/blocker" template("+${usracct.device}\n") ); }; filter f_ipt_recent { tags("secevt") and match("REJECT" value("secevt.verdict")); }; log { source (s_all); parser(pattern_db); filter(f_addtofirewall);destination(df_addtofirewall);}; ... which would be cool! Best regards, Valentijn

6 26

[Bug 121] New: After service syslog-ng reload, the first log don' t transfer to remote servicer via TLS
by bugzilla＠bugzilla.balabit.com 08 Feb '13

08 Feb '13

https://bugzilla.balabit.com/show_bug.cgi?id=121 Summary: After service syslog-ng reload, the first log don't transfer to remote servicer via TLS Product: syslog-ng Version: 3.0.x Platform: PC OS/Version: Linux Status: NEW Severity: critical Priority: unspecified Component: syslog-ng AssignedTo: bazsi(a)balabit.hu ReportedBy: yong.zheng(a)nsn.com Type of the Report: bug Estimated Hours: 0.0 The version:syslog-ng 3.0.5 options { flush_lines (0); time_reopen (10); log_fifo_size (1000); long_hostnames (off); use_dns (yes); use_fqdn (no); create_dirs (no); keep_hostname (yes); stats_freq (3600); }; I will upload the logs to remote syslog-ng server via TLS, I encounter a issue, On client: after reload the configure via below command: #service syslog-ng reload. I write a log: #logger -p local1.info "this is test" but I don't see the log to transfer to server immediately, I wait many minutes, still don't see. I must write the second log to triger that client send the 2 logs to server #logger -p local1.info "this is test 2" the 2 logs have transfered to server immediately successfully, I can see the 2 logs on server. It is 100% reproduced. Is this a known issue, or not ? Thanks. -- Configure bugmail: https://bugzilla.balabit.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching all bug changes.

1 6

[Bug 42] New: capabilities, chown, chmod
by bugzilla＠bugzilla.balabit.com 20 Dec '12

20 Dec '12

https://bugzilla.balabit.com/show_bug.cgi?id=42 Summary: capabilities, chown, chmod Product: syslog-ng Version: 3.0.x Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: unspecified Component: syslog-ng AssignedTo: bazsi(a)balabit.hu ReportedBy: zbyniu(a)pld-linux.org Type of the Report: bug Estimated Hours: 0.0 Let's take a look at syslog-ng-3.0.1/src/affile.c lines 60-83 1. CAP_SYS_ADMIN is needed only for /proc/kmsg, it is added w/o check 2. CAP_DAC_READ_SEARCH should be added only if open fail with errno 13 2a. CAP_DAC_OVERRIDE should be added only if open fail with errno 13 and with CAP_DAC_READ_SEARCH set 3. fchown needs CAP_CHOWN unconditionaly 4. fchmod needs CAP_FOWNER if file owner != euid (root here) 5. all caps should be restored summary: - CAP_SYS_ADMIN and CAP_DAC_OVERRIDE are set always even if unnecessary, and permanently - owner, group and perm doesn't work -- Configure bugmail: https://bugzilla.balabit.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching all bug changes.

2 10

kernel logging feature requests
by Patrick H. 01 Feb '12

01 Feb '12

While setting up a new server at home I've come across 2 feature requests that would be fairly nice to have (and not that hard to implement I would think). 1) When the 'kernel' flag is set on a file() source (like for reading /proc/kmsg), look for the printk time (eg "[ 1234.567890]") and calculate when the message would have occurred instead of just using when the line was read off the file. Basically check to see the number of seconds the system has been up, subtract the printk time, and then subtract that from current time. 2) I grab all kernel messages with priority of crit or higher and send it to the usertty() destination, but this destination doesnt support templates. It'd be nice to be able to define the template. I mostly just want to change the time format and remove the hostname (since these will only come from localhost on my setup).

4 8

patterndb & not describing the entire log message
by Christophe Brocas 22 Dec '11

22 Dec '11

Hello As I said before in other message, I currently work with patterndb to write a pattern file describing Windows log messages related to file system. These messages are sent to Syslog-NG through Snare software. Example : Aug 12 15:08:48 10.0.0.1 MSWinEventLog 1 Security 541 ven. août 12 15:08:46 201 4656 Microsoft-Windows-Security-Auditing D00000001\accout1 N/A Success Audit hostname.d00000001.subdom1.dom1.fr Système de fichiers Un handle vers un objet a été demandé. Sujet : ID de sécurité : S-1-5-21-514083418-1667312537-733111063-1179 Nom du compte : account1 Domaine du compte : D00000001 ID d’ouverture de session : 0x8e90858 Objet : Serveur de l’objet : Security Type d’objet : File Nom de l’objet : E:\xxx\yyyy\filename.doc.doc ID du handle : 0x2b0 Informations sur le processus : ID du processus : 0xeb0 Nom du processus : C:\Windows\System32\inetsrv\w3wp.exe Informations sur la demande d’accès : ID de la transaction : {00000000-0000-0000-0000-000000000000} Accès : DELETE SYNCHRONIZE ReadAttributes Masque d’accès : 0x110080 Privilèges utilisés pour les vérifications d’accès : - Nombre de SID restreints : 0 540 Question : That kind of messages are extremely long and some areas seem empty in my example but may be not in the future. Is there a way with pattern to not describe the entire message but only the relevant part for us (the equivalent of wildcard in regexp) ? Thanks Christophe ***************************************************** "Le contenu de ce courriel et ses éventuelles pièces jointes sont confidentiels. Ils s'adressent exclusivement à la personne destinataire. Si cet envoi ne vous est pas destiné, ou si vous l'avez reçu par erreur, et afin de ne pas violer le secret des correspondances, vous ne devez pas le transmettre à d'autres personnes ni le reproduire. Merci de le renvoyer à l'émetteur et de le détruire. Attention : L'organisme de l'émetteur du message ne pourra être tenu responsable de l'altération du présent courriel. Il appartient au destinataire de vérifier que les messages et pièces jointes reçus ne contiennent pas de virus. Les opinions contenues dans ce courriel et ses éventuelles pièces jointes sont celles de l'émetteur. Elles ne reflètent pas la position de l'organisme sauf s'il en est disposé autrement dans le présent courriel." ******************************************************

4 7

[Bug 111] New: RPM build errors: Installed (but unpackaged) file(s) found:
by bugzilla＠bugzilla.balabit.com 14 Nov '11

14 Nov '11

https://bugzilla.balabit.com/show_bug.cgi?id=111 Summary: RPM build errors: Installed (but unpackaged) file(s) found: Product: syslog-ng Version: 3.2.x Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: unspecified Component: packaging AssignedTo: folti(a)balabit.hu ReportedBy: owen.mann(a)interactivedata.com Type of the Report: bug Estimated Hours: 0.0 when attempting `rpmbuild -tb syslog-ng_3.2beta1.tar.gz`, the compilation completes, but: [...] Checking for unpackaged file(s): /usr/lib/rpm/check-files /var/tmp/syslog-ng-root error: Installed (but unpackaged) file(s) found: /bin/pdbtool /bin/update-patterndb /etc/syslog-ng/modules.conf /etc/syslog-ng/scl.conf /lib/libsyslog-ng.la /lib/libsyslog-ng.so /lib/libsyslog-ng.so.0 /lib/libsyslog-ng.so.0.0.0 /lib/syslog-ng/libaffile.la /lib/syslog-ng/libaffile.so /lib/syslog-ng/libafprog.la /lib/syslog-ng/libafprog.so /lib/syslog-ng/libafsocket-notls.la /lib/syslog-ng/libafsocket-notls.so /lib/syslog-ng/libafsocket-tls.la /lib/syslog-ng/libafsocket-tls.so /lib/syslog-ng/libafsocket.so /lib/syslog-ng/libafuser.la /lib/syslog-ng/libafuser.so /lib/syslog-ng/libbasicfuncs.la /lib/syslog-ng/libbasicfuncs.so /lib/syslog-ng/libconfgen.la /lib/syslog-ng/libconfgen.so /lib/syslog-ng/libconvertfuncs.la /lib/syslog-ng/libconvertfuncs.so /lib/syslog-ng/libcsvparser.la /lib/syslog-ng/libcsvparser.so /lib/syslog-ng/libdbparser.la /lib/syslog-ng/libdbparser.so /lib/syslog-ng/libdummy.la /lib/syslog-ng/libdummy.so /lib/syslog-ng/libsyslogformat.la /lib/syslog-ng/libsyslogformat.so /sbin/syslog-ng-ctl /share/include/scl/pacct/plugin.conf /share/include/scl/syslogconf/README /share/include/scl/syslogconf/convert-syslogconf.awk /share/include/scl/syslogconf/plugin.conf /share/include/scl/system/generate-system-source.sh /share/include/scl/system/plugin.conf /share/xsd/patterndb-1.xsd /share/xsd/patterndb-2.xsd /share/xsd/patterndb-3.xsd [...] If these files are added to the %files section in syslog-ng.spec, the build completes successfully. -- Configure bugmail: https://bugzilla.balabit.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching all bug changes.

1 9

Possible memleak in 3.3 HEAD
by Jakub Jankowski 12 Nov '11

12 Nov '11

Hi, I think I'm having a memleak in 3.3's HEAD. My production system, which receives ~300 messages/s throughout the day (so ~26.4 million messages daily) leaks ~700MB of memory per day. I did some investigation, here are the details: # syslog-ng -V syslog-ng 3.3.0beta2 Installer-Version: 3.3.0beta2 Revision: ssh+git://bazsi@git.balabit//var/scm/git/syslog-ng/syslog-ng-ose--mainline--3.3#no_branch#d6eea6a1130f8343b6600976d9f2319f9ed86c88 Compile-Date: Sep 9 2011 16:32:59 Default-Modules: affile,afprog,afsocket,afuser,basicfuncs,csvparser,dbparser,syslogformat,afsql Available-Modules: syslogformat,affile,afmongodb,basicfuncs,csvparser,dbparser,afsql,convertfuncs,afprog,confgen,afsocket,afuser,afsocket-tls,dummy Enable-Debug: off Enable-GProf: off Enable-Memtrace: off Enable-IPv6: on Enable-Spoof-Source: on Enable-TCP-Wrapper: on Enable-Linux-Caps: on Enable-Pcre: off # My build is patched with: - https://github.com/algernon/libmongo-client/commit/93c5cff05feec5d7f421acfd… (so it compiles on rhel5) - https://lists.balabit.hu/pipermail/syslog-ng/2011-April/016375.html (I need this feature) I don't think any of these is the reason, though. My test configuration is basically this: $ cat /etc/syslog-ng/syslog-ng.conf @version: 3.3 options { time_reap (1); log_fifo_size (100000); use_dns (no); use_fqdn (no); create_dirs (yes); keep_hostname (yes); log_msg_size(16384); owner ("root"); group ("root"); perm (0644); dir_owner ("root"); dir_group ("root"); dir_perm (0755); }; source s_www { syslog(ip(0.0.0.0) port(515) flags(no-multi-line) log_iw_size(6000) max-connections(30) tags("src.syslog", "src.port.515", "www", "www.accesslog")); }; filter f_svc_www_accesslog { tags("www.accesslog"); }; parser p_apache_split_vhost_access { csv-parser( columns("APACHE.ACCESS.VHOST", "APACHE.ACCESS.LOG_MESSAGE") flags(greedy) delimiters(" ") ); }; parser dbp_apache_split_vhost_access { db-parser( file("/etc/syslog-ng/pdb1.xml") ); }; destination d_access_split { file("/logs/$S_YEAR/$S_MONTH/$S_DAY/${APACHE.ACCESS.VHOST:-default}-access.log-${S_YEAR}${S_MONTH}${S_DAY}" template("${APACHE.ACCESS.LOG_MESSAGE}\n")); }; log { source(s_www); filter(f_svc_www_accesslog); parser(dbp_apache_split_vhost_access); destination(d_access_split); flags(final); }; $ Those two parsers (csv-parser and db-parser) are doing exactly the same, they split my Apache HTTPd access logs into two parts: first column (which is the vhost name) and the rest of the line. I have them both because I needed to know if that's a leak in any particular parser type. (Apparently not) I'm not very experienced with valgrind, but I did some tests, and what I've found is that the number of "definitely lost" blocks depends on the number of files created by d_access_split destination. So I took 10000 lines of some real httpd logs and did two tests: 1) replace real vhost names with completely random strings, to maximize number of files created by d_access_split 2) replace real vhost names with constant string, so my destination always writes to the same file. And this is what I got, using "G_SLICE=always-malloc valgrind --leak-check=full -v --show-reachable=yes --track-origins=yes --log-file=valgrind.log syslog-ng -F --no-caps -p /var/run/syslog-ng.pid" 1) random files: ==22416== 3,675,952 bytes in 9,989 blocks are definitely lost in loss record 578 of 578 ==22416== at 0x4021EC2: calloc (vg_replace_malloc.c:418) ==22416== by 0x40FF57D: g_malloc0 (in /lib/libglib-2.0.so.0.1200.3) ==22416== by 0x4052899: log_queue_fifo_new (logqueue-fifo.c:438) ==22416== by 0x4046057: log_dest_driver_acquire_queue_method (driver.c:153) ==22416== by 0x47081D1: affile_dw_init (driver.h:178) ==22416== by 0x4707C41: affile_dd_open_writer (logpipe.h:239) ==22416== by 0x405C258: main_loop_call (mainloop.c:145) ==22416== by 0x4707A6B: affile_dd_queue (affile.c:1127) ==22416== by 0x40451C8: log_dest_group_queue (logpipe.h:288) ==22416== by 0x404B9DD: log_multiplexer_queue (logpipe.h:288) ==22416== by 0x404EC16: log_parser_queue (logpipe.h:288) ==22416== by 0x4047473: log_filter_pipe_queue (logpipe.h:288) ==22416== ==22416== LEAK SUMMARY: ==22416== definitely lost: 3,676,156 bytes in 9,992 blocks ==22416== indirectly lost: 6,541 bytes in 56 blocks ==22416== possibly lost: 16,967 bytes in 117 blocks ==22416== still reachable: 76,908 bytes in 3,286 blocks ==22416== suppressed: 0 bytes in 0 blocks 2) single file (constant filename): ==19460== 368 bytes in 1 blocks are definitely lost in loss record 525 of 577 ==19460== at 0x4021EC2: calloc (vg_replace_malloc.c:418) ==19460== by 0x40FF57D: g_malloc0 (in /lib/libglib-2.0.so.0.1200.3) ==19460== by 0x4052899: log_queue_fifo_new (logqueue-fifo.c:438) ==19460== by 0x4046057: log_dest_driver_acquire_queue_method (driver.c:153) ==19460== by 0x47081D1: affile_dw_init (driver.h:178) ==19460== by 0x4707C41: affile_dd_open_writer (logpipe.h:239) ==19460== by 0x405C258: main_loop_call (mainloop.c:145) ==19460== by 0x4707A6B: affile_dd_queue (affile.c:1127) ==19460== by 0x40451C8: log_dest_group_queue (logpipe.h:288) ==19460== by 0x404B9DD: log_multiplexer_queue (logpipe.h:288) ==19460== by 0x404EC16: log_parser_queue (logpipe.h:288) ==19460== by 0x4047473: log_filter_pipe_queue (logpipe.h:288) (...) ==19460== LEAK SUMMARY: ==19460== definitely lost: 572 bytes in 4 blocks ==19460== indirectly lost: 6,541 bytes in 56 blocks ==19460== possibly lost: 12,919 bytes in 106 blocks ==19460== still reachable: 76,908 bytes in 3,286 blocks ==19460== suppressed: 0 bytes in 0 blocks Could anyone proficient in syslog-ng source code take a look on this? Can I do anything else to help you diagnose this? (I can provide complete valgrind output logs if this helps.) Also, I'll try to reproduce this on other versions later this week, unless someone fixes this earlier. :) Cheers, -- Jakub Jankowski|shasta@toxcorp.com|http://toxcorp.com/ GPG: FCBF F03D 9ADB B768 8B92 BB52 0341 9037 A875 942D

7 43

syslog-ng 3.2.4 MySQL connection loss Server has gone away
by Erik Maciejewski 12 Nov '11

12 Nov '11

Hello, I have been researching an issue with syslog-ng 3.2.4 (both the Linux glibc2.3.6 and platform independent compiled from source versions) and MySQL on CentOS 5.6 in which syslog-ng tries to insert a log messages using a dead TCP or unix socket connection to a MySQL database. The issue results in the max attempt to insert a log message and then the message subsequently dropped. This seems to occur regularly in predominately low message volume environments, but has the potential to affect all environments using a MySQL (or other db) destination. The root cause of the issue seems to be directly related to the health of the connection between syslog-ng and MySQL and can be affected by the "wait_timeout" value used by MySQL to kill off inactive connections. I want to provide justification for implementing a fix to syslog-ng as I believe manipulating a, possibly tuned, environment variable for MySQL would seem like the wrong approach to take for remedial action. I feel that many MySQL instances are implemented in shared environments and are themselves shared by many applications. New connections to a MySQL instance are generally regarded as low cost and, it would seem, more often than not never used in a persisted fashion when supporting distributed applications. That being said, many times the "wait_timeout" value is set very low so that the MySQL instance can effectively control the number of simultaneous connections. By taking somewhat of a black-box approach to the issue (I'm just starting to explore syslog-ng), I noticed that there doesn't seem to be any health checking of the TCP or unix socket connection in the SQL destination implementation after the initial connection is made (most of the db actions being abstracted away by libdbi). Even if a syslog-ng database thread is suspended and reactivated due to error, there does not seem to be any attempt to check the health of the database connection. I have implemented a fix in my compiled version that will check a database connection's health and attempt to reconnect upon discovering a dead connection. I would like to know the policies/procedures/best practices for submitting such issues and/or fixes for discussion, review, and implementation. If this is a known issue or I am just wrong, all the better! Thank you guys for keeping this project going and I'm looking forward to implementing syslog-ng in a production environment in the very near future! Regards, Erik Maciejewksi

2 7

Loganalyzer
by Peter Czanik 11 Nov '11

11 Nov '11

Hello, Recently I found many e-mail and forum posts looking for a web GUI for syslog-ng. One of them, Loganalyzer is getting more popular, as a completely free web application. There is just one catch, it was not designed for syslog-ng originally. So I did some research and tests, and using my instructions quite easy to get started with it. Of course Loganalyzer has some rough edges, but it works for me on a number of different systems. For a step by step tutorial to get started, read my blog at http://czanik.blogs.balabit.com/2011/02/loganalyzer/ Have a nice weekend! Bye, -- Peter Czanik (CzP) <czanik(a)balabit.hu> BalaBit IT Security / syslog-ng upstream http://czanik.blogs.balabit.com/

2 3

[Bug 96] New: pdbtool misleading error message
by bugzilla＠bugzilla.balabit.com 01 Nov '11

01 Nov '11

https://bugzilla.balabit.com/show_bug.cgi?id=96 Summary: pdbtool misleading error message Product: syslog-ng Version: 3.2.x Platform: PC OS/Version: Linux Status: NEW Severity: normal Priority: unspecified Component: syslog-ng AssignedTo: bazsi(a)balabit.hu ReportedBy: czanik(a)balabit.hu Type of the Report: --- Estimated Hours: 0.0 When trying to test a non-existent file with pdbtool it gives misleading error messages and even calls an external application before failing. czanik@linux-6y8u:~/patterndb> pdbtool test --validate uw-imapd.ceee uw-imapd.ceee: Unable to detect patterndb version, please write the <patterndb> tag on a single line warning: failed to load external entity "/usr/share/syslog-ng/xsd/patterndb-0.xsd" Schemas parser error : Failed to locate the main schema resource at '/usr/share/syslog-ng/xsd/patterndb-0.xsd'. WXS schema /usr/share/syslog-ng/xsd/patterndb-0.xsd failed to compile warning: failed to load external entity "uw-imapd.ceee" uw-imapd.ceee: xmllint returned an error, the executed command was: xmllint --noout --nonet --schema /usr/share/syslog-ng/xsd/patterndb-0.xsd uw-imapd.ceeeError opening classifier configuration file; filename='uw-imapd.ceee', error='No such file or directory (2)' I think, that the proper behavior would be, that pdbtool detect that the file is nonexistent, prints a message about it (without any other misleading messages) and quits without calling external apps. -- Configure bugmail: https://bugzilla.balabit.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching all bug changes.

1 2