Segfault on ubuntu 12.04.4 with monit 5.8

Issue #37 resolved
Vincent Spaa created an issue

On our production system we're getting the following log line: localhost kernel: [350688.719741] monit[32633]: segfault at 19 ip 00007f9dc94f0714 sp 00007fff7f06a8d0 error 4 in libc-2.15.so[7f9dc9471000+1b5000] After which the monit daemon is dead.

This doesn't seem to get triggered by any particular reason, at least not for as far as the logs tell us. We use monit to monitor a few game servers, a simple bash script and a voice server.

It's a x64 version of Ubuntu 12.04.4 and this is the uname output: 3.8.0-29-generic #42~precise1-Ubuntu SMP Wed Aug 14 16:19:23 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux And as the title suggests, we're using monit 5.8 which was installed with apt-get.

Attached you'll find our monit config.

Appologies in advance if this is something we're doing wrong rather than a bug manifesting itself.

Comments (9)

  1. Pawel Kubica

    I also have problems with Monit 5.8 (downloaded x86 binary from site) on Gentoo fresh install:

    glibc detected monit: double free or corruption (!prev): 0x096ae0b0 *** ======= Backtrace: ========= /lib/libc.so.6[0xb75f1fc4] /lib/libc.so.6[0xb75f3693] /lib/libc.so.6(cfree+0x6d)[0xb75f66ad] monit[0x8084291] monit[0x806249f] monit[0x806e715] monit[0x8058860] monit[0x80594e1] monit[0x805a983] monit[0x805b3e7] monit[0x807135d] monit[0x807312b] monit[0x8056fb5] monit[0x8057b95] /lib/libc.so.6(__libc_start_main+0xe5)[0xb759ca65] monit[0x804b5e1] ======= Memory map: ======== 08048000-081e9000 r-xp 00000000 08:03 8605 /usr/local/bin/monit 081e9000-081f2000 rw-p 001a0000 08:03 8605 /usr/local/bin/monit 081f2000-081f7000 rw-p 00000000 00:00 0 09676000-096cf000 rw-p 00000000 00:00 0 [heap] b6400000-b6421000 rw-p 00000000 00:00 0 b6421000-b6500000 ---p 00000000 00:00 0 b6552000-b655e000 r-xp 00000000 08:03 245836 /usr/lib/gcc/i686-pc-linux-gnu/4.3.4/libgcc_s.so.1 b655e000-b655f000 r--p 0000b000 08:03 245836 /usr/lib/gcc/i686-pc-linux-gnu/4.3.4/libgcc_s.so.1 b655f000-b6560000 rw-p 0000c000 08:03 245836 /usr/lib/gcc/i686-pc-linux-gnu/4.3.4/libgcc_s.so.1 b6564000-b6565000 ---p 00000000 00:00 0 b6565000-b6d65000 rw-p 00000000 00:00 0 b6d65000-b6d66000 ---p 00000000 00:00 0 b6d66000-b7566000 rw-p 00000000 00:00 0 b7566000-b7570000 r-xp 00000000 08:03 147700 /lib/libnss_files-2.10.1.so b7570000-b7571000 r--p 00009000 08:03 147700 /lib/libnss_files-2.10.1.so b7571000-b7572000 rw-p 0000a000 08:03 147700 /lib/libnss_files-2.10.1.so b7572000-b757b000 r-xp 00000000 08:03 147702 /lib/libnss_nis-2.10.1.so b757b000-b757c000 r--p 00008000 08:03 147702 /lib/libnss_nis-2.10.1.so b757c000-b757d000 rw-p 00009000 08:03 147702 /lib/libnss_nis-2.10.1.so b757d000-b7583000 r-xp 00000000 08:03 147698 /lib/libnss_compat-2.10.1.so b7583000-b7584000 r--p 00006000 08:03 147698 /lib/libnss_compat-2.10.1.so b7584000-b7585000 rw-p 00007000 08:03 147698 /lib/libnss_compat-2.10.1.so b7585000-b7586000 rw-p 00000000 00:00 0 b7586000-b76c8000 r-xp 00000000 08:03 147681 /lib/libc-2.10.1.so b76c8000-b76ca000 r--p 00142000 08:03 147681 /lib/libc-2.10.1.so b76ca000-b76cb000 rw-p 00144000 08:03 147681 /lib/libc-2.10.1.so b76cb000-b76cf000 rw-p 00000000 00:00 0

    This happens when Monit try restart process (openvpn) every cycle:

    check process openvpn matching "/usr/sbin/openvpn"
            restart = "/etc/init.d/openvpn restart"
            if does not exist 1 times within 1 cycles then restart else if succeeded 2 times within 2 cycles then alert
            if 6 restarts within 6 cycles then alert
    

    I can attach coredump (17M) if it helps...

  2. Tildeslash repo owner

    Hello Pawel,

    we're unable to replicate the problem with the given configuration.

    Please can you send the coredump + full monit conifguration file (obfuscate the credentials for if any) + monit log file to support@mmonit.com?

    Best regards, The Monit team

  3. Vincent Spaa reporter

    Many thanks for the quick fix. I will post back if we keep on running into the issue.

  4. Log in to comment