Results 1 to 15 of 15
  1. #1
    Join Date
    Oct 2003
    Location
    Georgetown, Ontario
    Posts
    1,771

    rndc/named Seg Fault on kernel 2.4.26 w/ RHEL

    We just upgraded our test RedHat Enterprise Server (Dual Xeon) to the 2.4.26 kernel. Everything works, except for bind/rndc.

    When I try to restart named it show this.

    Stopping named: /etc/init.d/named: line 112: 31226 Segmentation fault /usr/sbin/rndc stop

    When ever I try to do anything with rndc, all it gives back is "Segmentation Fault".


    Does anyone have any ideas?
    ·· Repeat after me... ProSupport is the best... Prosupport is... ··
    ProSupport Host Support System - OUT NOW! Grab a copy yourself and see what the hype is about!
    VertiHost Inc. - We run a quality business. Do you?

  2. #2
    Join Date
    Apr 2003
    Location
    UK
    Posts
    2,569
    what happens when you strace it ?

    strace rndc stop

  3. #3
    Join Date
    Oct 2003
    Location
    Georgetown, Ontario
    Posts
    1,771
    root@cp06 [~]# strace rndc stop
    execve("/usr/local/sbin/rndc", ["rndc", "stop"], [/* 27 vars */]) = 0
    uname({sys="Linux", node="cp06.vertihost.com", ...}) = 0
    brk(0) = 0x80778d0
    open("/etc/ld.so.preload", O_RDONLY) = -1 ENOENT (No such file or directory)
    open("/etc/ld.so.cache", O_RDONLY) = 3
    fstat64(3, {st_mode=S_IFREG|0644, st_size=36681, ...}) = 0
    old_mmap(NULL, 36681, PROT_READ, MAP_PRIVATE, 3, 0) = 0x40016000
    close(3) = 0
    open("/lib/libnsl.so.1", O_RDONLY) = 3
    read(3, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0000<\0\000"..., 512) = 512
    fstat64(3, {st_mode=S_IFREG|0755, st_size=91368, ...}) = 0
    old_mmap(NULL, 85184, PROT_READ|PROT_EXEC, MAP_PRIVATE, 3, 0) = 0x4001f000
    old_mmap(0x40031000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 3, 0x11000) = 0x40031000
    old_mmap(0x40032000, 7360, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x40032000
    close(3) = 0
    open("/lib/tls/libpthread.so.0", O_RDONLY) = 3
    read(3, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0\300F\0"..., 512) = 512
    fstat64(3, {st_mode=S_IFREG|0755, st_size=97708, ...}) = 0
    old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40034000
    old_mmap(NULL, 64852, PROT_READ|PROT_EXEC, MAP_PRIVATE, 3, 0) = 0x40035000
    old_mmap(0x40042000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 3, 0xc000) = 0x40042000
    old_mmap(0x40043000, 7508, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x40043000
    close(3) = 0
    open("/lib/tls/libc.so.6", O_RDONLY) = 3
    read(3, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0000X\1\000"..., 512) = 512
    fstat64(3, {st_mode=S_IFREG|0755, st_size=1567528, ...}) = 0
    old_mmap(NULL, 1276748, PROT_READ|PROT_EXEC, MAP_PRIVATE, 3, 0) = 0x40045000
    old_mmap(0x40177000, 16384, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 3, 0x131000) = 0x40177000
    old_mmap(0x4017b000, 6988, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x4017b000
    close(3) = 0
    old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x4017d000
    set_thread_area({entry_number:-1 -> -1, base_addr:0x4017d080, limit:1048575, seg_32bit:1, contents:0, read_exec_only:0, limit_in_pages:1, seg_not_present:0, useable:1}) = -1 ENOSYS (Function not implemented)
    modify_ldt(1, {entry_number:0, base_addr:0x4017d080, limit:1048575, seg_32bit:1, contents:0, read_exec_only:0, limit_in_pages:1, seg_not_present:0, useable:1}, 16) = 0
    munmap(0x40016000, 36681) = 0
    set_tid_address(0x4017d0c8) = -1 ENOSYS (Function not implemented)
    rt_sigaction(SIGRTMIN, {0x40039620, [], SA_RESTORER|SA_SIGINFO, 0x4003fe40}, NULL, 8) = 0
    rt_sigprocmask(SIG_UNBLOCK, [RTMIN], NULL, 8) = 0
    getrlimit(RLIMIT_STACK, {rlim_cur=8192*1024, rlim_max=8192*1024}) = 0
    rt_sigaction(SIGINT, {0x805e27c, ~[RTMIN], SA_RESTORER, 0x4003fe48}, NULL, 8) = 0
    rt_sigaction(SIGTERM, {0x805e27c, ~[RTMIN], SA_RESTORER, 0x4003fe48}, NULL, 8) = 0
    rt_sigaction(SIGPIPE, {SIG_IGN}, NULL, 8) = 0
    rt_sigaction(SIGHUP, {SIG_DFL}, NULL, 8) = 0
    rt_sigprocmask(SIG_BLOCK, [HUP INT TERM], NULL, 8) = 0
    getpid() = 5697
    time(NULL) = 1083324661
    futex(0x8075774, FUTEX_WAKE, 2147483647) = -1 ENOSYS (Function not implemented)
    brk(0) = 0x80778d0
    brk(0x80988d0) = 0x80988d0
    brk(0) = 0x80988d0
    brk(0x8099000) = 0x8099000
    mmap2(NULL, 139264, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x4017e000
    pipe([3, 4]) = 0
    fcntl64(3, F_GETFL) = 0 (flags O_RDONLY)
    fcntl64(3, F_SETFL, O_RDONLY|O_NONBLOCK) = 0
    mmap2(NULL, 8392704, PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x401a0000
    mprotect(0x401a0000, 4096, PROT_NONE) = 0
    clone(child_stack=0x409a0b08, flags=CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHAND|CLONE_THREAD|CLONE_SYSVSEM|CLONE_SETTLS|CLONE_PARENT_SETTID|CLONE_CHILD_CLEARTID|CLONE_DETACHED, parent_tidptr=0x409a0bf8, {entry_number:0, base_addr:0x409a0bb0, limit:1048575, seg_32bit:1, contents:0, read_exec_only:0, limit_in_pages:1, seg_not_present:0, useable:1}, child_tidptr=0x409a0bf8) = 5698
    --- SIGSEGV (Segmentation fault) @ 0 (0) ---
    +++ killed by SIGSEGV +++
    ·· Repeat after me... ProSupport is the best... Prosupport is... ··
    ProSupport Host Support System - OUT NOW! Grab a copy yourself and see what the hype is about!
    VertiHost Inc. - We run a quality business. Do you?

  4. #4
    Join Date
    Apr 2003
    Location
    UK
    Posts
    2,569
    hmm thats not helpful (as far as my eyes can see!)

    anything in logs? or grsec in your kernel ?

  5. #5
    Join Date
    Oct 2003
    Location
    Georgetown, Ontario
    Posts
    1,771
    no grsec. Its just a stock 2.4.26 kernel compiled from source. Bind works with the stock RHEL kernel that comes preinstalled, just not this compiled version.

    Ill check the logs and get back.
    ·· Repeat after me... ProSupport is the best... Prosupport is... ··
    ProSupport Host Support System - OUT NOW! Grab a copy yourself and see what the hype is about!
    VertiHost Inc. - We run a quality business. Do you?

  6. #6
    Your bind as well as some other apps will segfault as RHEL has implemented new Native POSIX Thread Library which is not compatible with 2.4.X kernels.

  7. #7
    Join Date
    Oct 2003
    Location
    Georgetown, Ontario
    Posts
    1,771
    so what should I do?
    ·· Repeat after me... ProSupport is the best... Prosupport is... ··
    ProSupport Host Support System - OUT NOW! Grab a copy yourself and see what the hype is about!
    VertiHost Inc. - We run a quality business. Do you?

  8. #8
    Join Date
    Jul 2003
    Location
    Nothing but, net
    Posts
    2,064
    Originally posted by cMark
    so what should I do?
    http://www.freebsd.org/

    Upgrade of course..

  9. #9
    Join Date
    Oct 2003
    Location
    Georgetown, Ontario
    Posts
    1,771
    ha, other than that.....
    ·· Repeat after me... ProSupport is the best... Prosupport is... ··
    ProSupport Host Support System - OUT NOW! Grab a copy yourself and see what the hype is about!
    VertiHost Inc. - We run a quality business. Do you?

  10. #10
    Join Date
    Mar 2003
    Location
    California USA
    Posts
    13,681
    this is because all kernels but rhe kernels and 2.6 do not have nptl patching. It can be done to work even with grsecurity =)
    Steven Ciaburri | Industry's Best Server Management - Rack911.com
    Software Auditing - 400+ Vulnerabilities Found - Quote @ https://www.RACK911Labs.com
    Fully Managed Dedicated Servers (Las Vegas, New York City, & Amsterdam) (AS62710)
    FreeBSD & Linux Server Management, Security Auditing, Server Optimization, PCI Compliance

  11. #11
    Join Date
    Oct 2003
    Location
    Georgetown, Ontario
    Posts
    1,771
    me and choon finally got it all to worked out with nptl patching.


    If you are using Red Hat Enterprise Linux ES release 3 (Taroon Update 1), then take a look at this script and follow the instruction or else your will get Segfaults with bind and other programs cannot be started.

    http://choonhost.net/nptl-profile.txt
    Last edited by Pheaton; 04-30-2004 at 08:41 PM.
    ·· Repeat after me... ProSupport is the best... Prosupport is... ··
    ProSupport Host Support System - OUT NOW! Grab a copy yourself and see what the hype is about!
    VertiHost Inc. - We run a quality business. Do you?

  12. #12
    Join Date
    Mar 2003
    Location
    California USA
    Posts
    13,681
    yep me and choon discussed this the other day
    Steven Ciaburri | Industry's Best Server Management - Rack911.com
    Software Auditing - 400+ Vulnerabilities Found - Quote @ https://www.RACK911Labs.com
    Fully Managed Dedicated Servers (Las Vegas, New York City, & Amsterdam) (AS62710)
    FreeBSD & Linux Server Management, Security Auditing, Server Optimization, PCI Compliance

  13. #13
    Join Date
    Apr 2002
    Location
    Houston, Texas, U.S.A.
    Posts
    44

    Re: rndc/named Seg Fault on kernel 2.4.26 w/ RHEL

    Originally posted by Pheaton
    We just upgraded our test RedHat Enterprise Server (Dual Xeon) to the 2.4.26 kernel. Everything works, except for bind/rndc.

    When I try to restart named it show this.

    When ever I try to do anything with rndc, all it gives back is "Segmentation Fault".

    Does anyone have any ideas?
    Moving the tls directory seems to have resolved it, at least for me:

    mv /lib/tls/ /lib/tls.disable

    -edit-

    Heh, just now read what was in the nptl script from choon. Does the same thing.
    Last edited by Rooter; 08-22-2004 at 02:42 AM.
    Test, testing, 1, 2, 3... :]

  14. #14
    Join Date
    Mar 2003
    Location
    California USA
    Posts
    13,681
    adding

    export LD_ASSUME_KERNEL=2.2.5

    to the bottom of /etc/profile is a better idea then moving /lib/tls
    Steven Ciaburri | Industry's Best Server Management - Rack911.com
    Software Auditing - 400+ Vulnerabilities Found - Quote @ https://www.RACK911Labs.com
    Fully Managed Dedicated Servers (Las Vegas, New York City, & Amsterdam) (AS62710)
    FreeBSD & Linux Server Management, Security Auditing, Server Optimization, PCI Compliance

  15. #15
    Join Date
    Apr 2002
    Location
    Houston, Texas, U.S.A.
    Posts
    44
    Originally posted by thelinuxguy
    adding

    export LD_ASSUME_KERNEL=2.2.5

    to the bottom of /etc/profile is a better idea then moving /lib/tls
    Thanks for the tip, I wasn't aware of this.
    Test, testing, 1, 2, 3... :]

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •