Looks like Daniel Cid might have fixed this: https://bitbucket.org/dcid/ossec-hids/changeset/8cc93c407d69
On Wed, Aug 22, 2012 at 7:54 AM, dan (ddp) <[email protected]> wrote: > On Fri, Aug 17, 2012 at 7:56 PM, Jim <[email protected]> wrote: >> Dan, >> >> Here is the backtrace from GDB, but I am not sure that tells much more than >> mdb had? >> > > It's a tool I'm more familiar with. I don't get much of an opportunity > to use niche systems these days. > > I'd consider tossing the following into log.c before the fprintf to > see if it gives anymore info; > > merror(_fflog, > "%d %s %02d %s %s%s%s %s %s %s:%s->%s:%s\n", > lf->year, > lf->mon, > lf->day, > lf->hour, > lf->hostname != lf->location?lf->hostname:"", > lf->hostname != lf->location?"->":"", > lf->location, > lf->action, > lf->protocol, > lf->srcip, > lf->srcport, > lf->dstip, > lf->dstport); > > > Also, do you have any rules or anything looking at the firewall logs? > > Hopefully one of the actual developers will take a look at this and > just know what's going on. > >> Program terminated with signal 11, Segmentation fault. >> #0 0xfed95b9c in strlen () from /usr/lib/libc.so.1 >> (gdb) bt >> #0 0xfed95b9c in strlen () from /usr/lib/libc.so.1 >> #1 0xfedf0f82 in _ndoprnt () from /usr/lib/libc.so.1 >> #2 0xfedf3b09 in fprintf () from /usr/lib/libc.so.1 >> #3 0x08076d0d in FW_Log (lf=0x81c0558) at log.c:261 >> #4 0x08057530 in OS_ReadMSG (m_queue=6) at analysisd.c:860 >> #5 0x08056f45 in main (argc=1, argv=0x8047dcc) at analysisd.c:527 >> >> Any idea what is causing this? >> >> Thanks, >> >> >> --JIM >> >> On Thursday, August 16, 2012 8:54:43 AM UTC-4, dan (ddpbsd) wrote: >>> >>> On Thu, Aug 16, 2012 at 1:12 AM, Jim <[email protected]> wrote: >>> > Hello, >>> > >>> > Any further thoughts on fixing this core dump problem? >>> > >>> > Thanks, >>> > >>> > >>> > --JIM >>> > >>> >>> Is there any chance you can run it in gdb? >>> >>> gdb /var/ossec/bin/ossec-analysisd >>> set follow-fork-mode child >>> run >>> *CRASH* >>> bt >>> >>> >>> There are probably other things you should be running in addition to >>> bt, but I don't know it all that well. >>> >>> > >>> > On Monday, August 13, 2012 7:41:39 PM UTC-4, Jim wrote: >>> >> >>> >> Here are the logs from the ossec.log, which was running in debug. >>> >> Which >>> >> reports until you can see analysisd core dumps. IPs and hostnames have >>> >> been >>> >> changed from the original... >>> >> >>> >> 2012/08/12 15:07:09 ossec-logcollector: DEBUG: Reading syslog message: >>> >> 'Aug 12 15:07:08 casrv9-hidn.local1 ipmon[123]: 15:07:07.674864 e1000g0 >>> >> @0:63 p 192.168.2.81,46618 -> 192.168.2.205,994 PR tcp len 20 142 -AP >>> >> IN' >>> >> 2012/08/12 15:07:09 ossec-logcollector: DEBUG: Reading syslog message: >>> >> 'Aug 12 15:07:08 casrv9-hidn.local1 ipmon[123]: 15:07:07.675807 e1000g0 >>> >> @0:63 p 192.168.2.81,46618 -> 192.168.2.205,994 PR tcp len 20 52 -A IN' >>> >> 2012/08/12 15:07:09 ossec-logcollector: DEBUG: Reading syslog message: >>> >> 'Aug 12 15:07:08 batch.local1 batch kernel: VFS: busy inodes on changed >>> >> media or resized disk hdb' >>> >> 2012/08/12 15:07:09 ossec-logcollector: DEBUG: Reading syslog message: >>> >> 'Aug 12 15:07:08 n58.local1 ntpd[2816]: kernel time sync enabled 0001' >>> >> 2012/08/12 15:07:09 ossec-logcollector: DEBUG: Reading syslog message: >>> >> 'Aug 12 15:07:09 n44.local1 smartd[2942]: Device: /dev/sda, 661 >>> >> Currently >>> >> unreadable (pending) sectors ' >>> >> 2012/08/12 15:07:11 ossec-logcollector: DEBUG: Reading syslog message: >>> >> 'Aug 12 15:07:10 FW1.local1 kernel: ACCEPTED IN=eth0 OUT=eth4 >>> >> SRC=192.168.1.200 DST=172.16.4.34 LEN=84 TOS=0x00 PREC=0x00 TTL=252 >>> >> ID=27140 >>> >> PROTO=ICMP TYPE=8 CODE=0 ID=16044 SEQ=0 ' >>> >> 2012/08/12 15:07:11 ossec-logcollector: DEBUG: Reading syslog message: >>> >> 'Aug 12 15:07:10 batch.local1 batch kernel: VFS: busy inodes on changed >>> >> media or resized disk hdb' >>> >> 2012/08/12 15:07:11 ossec-logcollector: DEBUG: Reading syslog message: >>> >> 'Aug 12 15:07:11 casrv9-hidn.local1 ipmon[123]: 15:07:10.875337 e1000g0 >>> >> @0:60 p 192.168.4.52,49168 -> 192.168.2.205,6667 PR tcp len 20 72 -AP >>> >> IN' >>> >> 2012/08/12 15:07:11 ossec-logcollector: DEBUG: Reading syslog message: >>> >> 'Aug 12 15:07:11 casrv9-hidn.local1 ipmon[123]: 15:07:10.876005 e1000g0 >>> >> @0:60 p 192.168.4.52,49168 -> 192.168.2.205,6667 PR tcp len 20 52 -A >>> >> IN' >>> >> 2012/08/12 15:07:11 ossec-logcollector: DEBUG: Reading syslog message: >>> >> 'Aug 12 15:07:11 casrv9-hidn.local1 ipmon[123]: 15:07:11.071610 e1000g0 >>> >> @0:63 p 192.168.2.127,56476 -> 192.168.2.205,994 PR tcp len 20 142 -AP >>> >> IN' >>> >> 2012/08/12 15:07:11 ossec-logcollector: DEBUG: Reading syslog message: >>> >> 'Aug 12 15:07:11 casrv9-hidn.local1 ipmon[123]: 15:07:11.074955 e1000g0 >>> >> @0:63 p 192.168.2.127,56476 -> 192.168.2.205,994 PR tcp len 20 52 -A >>> >> IN' >>> >> 2012/08/12 15:07:11 ossec-logcollector: DEBUG: Reading syslog message: >>> >> 'Aug 12 15:07:11 casrv9-hidn.local1 ipmon[123]: 15:07:11.140966 e1000g0 >>> >> @0:63 p 192.168.2.81,46618 -> 192.168.2.205,994 PR tcp len 20 126 -AP >>> >> IN' >>> >> 2012/08/12 15:07:11 ossec-logcollector: DEBUG: Reading syslog message: >>> >> 'Aug 12 15:07:11 casrv9-hidn.local1 ipmon[123]: 15:07:11.142122 e1000g0 >>> >> @0:63 p 192.168.2.81,46618 -> 192.168.2.205,994 PR tcp len 20 52 -A IN' >>> >> 2012/08/12 15:07:13 ossec-logcollector: DEBUG: Reading syslog message: >>> >> 'Aug 12 15:07:12 loghost unix: NOTICE: core_log: ossec-analysisd[713] >>> >> core >>> >> dumped: /var/cores/core.ossec-analysisd.713' >>> >> >>> >> >>> >> And here is the firewall log from ossec. I can see that these messages >>> >> stop before the above messages: >>> >> >>> >> 2012 Aug 12 15:06:57 casrv9-hidn.local1->/var/adm/systemlog ALLOW tcp >>> >> 192.168.4.142:45755->192.168.2.205:994 >>> >> 2012 Aug 12 15:06:59 casrv9-hidn.local1->/var/adm/systemlog ALLOW tcp >>> >> 192.168.4.142:45755->192.168.2.205:994 >>> >> 2012 Aug 12 15:06:59 casrv9-hidn.local1->/var/adm/systemlog ALLOW tcp >>> >> 192.168.4.127:58951->192.168.2.205:994 >>> >> 2012 Aug 12 15:06:59 casrv9-hidn.local1->/var/adm/systemlog ALLOW tcp >>> >> 192.168.4.127:58951->192.168.2.205:994 >>> >> 2012 Aug 12 15:07:01 casrv9-hidn.local1->/var/adm/systemlog ALLOW tcp >>> >> 192.168.15.70:53959->192.168.2.205:994 >>> >> 2012 Aug 12 15:07:01 casrv9-hidn.local1->/var/adm/systemlog ALLOW tcp >>> >> 192.168.15.70:53959->192.168.2.205:994 >>> >> 2012 Aug 12 15:07:05 casrv9-hidn.local1->/var/adm/systemlog ALLOW tcp >>> >> 192.168.4.61:49195->192.168.2.205:6667 >>> >> 2012 Aug 12 15:07:05 casrv9-hidn.local1->/var/adm/systemlog ALLOW tcp >>> >> 192.168.4.61:49195->192.168.2.205:6667 >>> >> 2012 Aug 12 15:07:09 casrv9-hidn.local1->/var/adm/systemlog ALLOW tcp >>> >> 192.168.2.81:46618->192.168.2.205:994 >>> >> 2012 Aug 12 15:07:09 casrv9-hidn.local1->/var/adm/systemlog ALLOW tcp >>> >> 192.168.2.81:46618->192.168.2.205:994 >>> >> >>> >> Last couple of lines in firewall log match the first couple of lines in >>> >> the ossec.log snippet. I am guessing things went bad after the last >>> >> entry >>> >> reported in the firewall log? >>> >> >>> >> >>> >> --JIM >>> >> >>> >> >>> >> On Monday, August 13, 2012 3:03:53 PM UTC-4, JB wrote: >>> >>> >>> >>> Seg fault occurred at ./analysisd/alerts/log.c around line 261: >>> >>> fprintf(_fflog, >>> >>> "%d %s %02d %s %s%s%s %s %s %s:%s->%s:%s\n", >>> >>> lf->year, >>> >>> lf->mon, >>> >>> lf->day, >>> >>> lf->hour, >>> >>> lf->hostname != lf->location?lf->hostname:"", >>> >>> lf->hostname != lf->location?"->":"", >>> >>> lf->location, >>> >>> lf->action, >>> >>> lf->protocol, >>> >>> lf->srcip, >>> >>> lf->srcport, >>> >>> lf->dstip, >>> >>> lf->dstport); >>> >>> >>> >>> Could you provide some Solaris firewall raw log entries prior to the >>> >>> time >>> >>> of this core dump? >>> >>> Also, provide same OSSEC firewall log entries (e.g., >>> >>> /var/ossec/logs/fiewwall/2012/Aug/oosec-firewall-12.log). >>> >>> It seems one of the lf-> fields was null but I do not know which one >>> >>> yet. >>> >>> >>> >>> On Sunday, August 12, 2012 12:19:23 PM UTC-7, Jim wrote: >>> >>>> >>> >>>> Hi, >>> >>>> >>> >>>> I am trying to get OSSEC version 2.6 running on one of our Solaris 10 >>> >>>> loghosts but I am running into a problem where analysisd cores after >>> >>>> about a >>> >>>> minute. It will actually report some e-mail alerts for a minute or >>> >>>> so >>> >>>> before analysisd cores. Below is an MDeBug of the core file: >>> >>>> >>> >>>> >>> >>>> >>> >>>> >>> >>>> ****************************************************************************** >>> >>>> Application core Dump Analysis Output MDeBug >>> >>>> Rev >>> >>>> 1.0 >>> >>>> Sun Aug 12 15:13:15 EDT 2012 Files: >>> >>>> /var/ossec/bin/ossec-analysisd core.ossec-analysisd.713 >>> >>>> >>> >>>> >>> >>>> ****************************************************************************** >>> >>>> >>> >>>> >>> >>>> >>> >>>> ** Core file status ** >>> >>>> ------------------------ >>> >>>> debugging core file of ossec-analysisd (32-bit) from ex >>> >>>> file: /var/ossec/bin/ossec-analysisd >>> >>>> initial argv: /var/ossec/bin/ossec-analysisd -d -d >>> >>>> threading model: multi-threaded >>> >>>> status: process terminated by SIGSEGV (Segmentation Fault) >>> >>>> >>> >>>> >>> >>>> ** Thread stack($c) ** >>> >>>> ---------------------- >>> >>>> libc.so.1`strlen+0xc(8095e28, 8046478, 80bbc10, 0) >>> >>>> libc.so.1`fprintf+0x99(80bbc10, 8095e08, 7dc, 81b31f2, c, 81b31e8) >>> >>>> FW_Log+0x38e(81b3178, 81b3178, 8046500, 805763c) >>> >>>> OS_ReadMSG+0x5d0(4, 808d48b, 2c9, 808d46b) >>> >>>> main+0x98a(3, 8047db0, 8047dc0) >>> >>>> _start+0x80(3, 8047e74, 8047e93, 8047e96, 0, 8047e99) >>> >>>> >>> >>>> >>> >>>> ** Shared objects ** >>> >>>> ---------------------- >>> >>>> BASE LIMIT SIZE NAME >>> >>>> 8050000 80a8000 58000 /var/ossec/bin/ossec-analysisd >>> >>>> fef90000 fef9b000 b000 /lib/libsocket.so.1 >>> >>>> feef0000 fef71000 81000 /lib/libnsl.so.1 >>> >>>> feea0000 feed6000 36000 /lib/libresolv.so.2 >>> >>>> fed70000 fee7e000 10e000 /lib/libc.so.1 >>> >>>> fefc3000 fefeb000 28000 /lib/ld.so.1 >>> >>>> >>> >>>> >>> >>>> Thread stack for MT app >>> >>>> ------------------------ >>> >>>> stack pointer for thread 1: 8046440 >>> >>>> [ 08046440 libc.so.1`strlen+0xc() ] >>> >>>> 08046468 libc.so.1`fprintf+0x99() >>> >>>> 080464c8 FW_Log+0x38e() >>> >>>> 08047d28 OS_ReadMSG+0x5d0() >>> >>>> 08047d8c main+0x98a() >>> >>>> 08047da4 _start+0x80() >>> >>>> >>> >>>> >>> >>>> Then a disassembly of FW_log. Let me know if you need me to dig >>> >>>> deeper >>> >>>> here... >>> >>>> >>> >>>> bash-3.00# mdb /var/ossec/bin/ossec-analysisd >>> >>>> core.ossec-analysisd.713 >>> >>>> Loading modules: [ libc.so.1 ld.so.1 ] >>> >>>> > ::stack >>> >>>> libc.so.1`strlen+0xc(8095e28, 8046478, 80bbc10, 0) >>> >>>> libc.so.1`fprintf+0x99(80bbc10, 8095e08, 7dc, 81b31f2, c, 81b31e8) >>> >>>> FW_Log+0x38e(81b3178, 81b3178, 8046500, 805763c) >>> >>>> OS_ReadMSG+0x5d0(4, 808d48b, 2c9, 808d46b) >>> >>>> main+0x98a(3, 8047db0, 8047dc0) >>> >>>> _start+0x80(3, 8047e74, 8047e93, 8047e96, 0, 8047e99) >>> >>>> >>> >>>> > FW_Log+0x38e::dis >>> >>>> FW_Log+0x36b: movl 0x8(%ebp),%eax >>> >>>> FW_Log+0x36e: pushl 0x68(%eax) >>> >>>> FW_Log+0x371: movl 0x8(%ebp),%eax >>> >>>> FW_Log+0x374: addl $0x7a,%eax >>> >>>> FW_Log+0x377: pushl %eax >>> >>>> FW_Log+0x378: movl 0x8(%ebp),%eax >>> >>>> FW_Log+0x37b: pushl 0x6c(%eax) >>> >>>> FW_Log+0x37e: pushl $0x8095e08 >>> >>>> FW_Log+0x383: pushl 0x80c068c >>> >>>> FW_Log+0x389: call -0x20f6c >>> >>>> <PLT=libc.so.1`fprintf> >>> >>>> FW_Log+0x38e: addl $0x40,%esp >>> >>>> FW_Log+0x391: subl $0xc,%esp >>> >>>> FW_Log+0x394: pushl 0x80c068c >>> >>>> FW_Log+0x39a: call -0x20e5d >>> >>>> <PLT=libc.so.1`fflush> >>> >>>> FW_Log+0x39f: addl $0x10,%esp >>> >>>> FW_Log+0x3a2: movl $0x1,-0x8(%ebp) >>> >>>> FW_Log+0x3a9: movl -0x8(%ebp),%eax >>> >>>> FW_Log+0x3ac: movl -0x4(%ebp),%ebx >>> >>>> FW_Log+0x3af: leave >>> >>>> FW_Log+0x3b0: ret >>> >>>> mknod: pushl %ebp >>> >>>> > ::quit >>> >>>> >>> >>>> >>> >>>> >>> >>>> --JIM
