/usr/src/blcr-0.8.2/doc/html/FAQ.html

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
    <title>BLCR Frequently Asked Questions (for version 0.8.2)</title>
 <meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1">
</head>
<body bgcolor="#ffffff">

<h1>BLCR Frequently Asked Questions (for version 0.8.2)</h1>


<h3>General Questions</h3>
<ul>
    <li><a href="#whatisblcr">What is BLCR?</a>
    <li><a href="#sigstop">How is checkpoint/restart different than SIGSTOP/SIGCONT?</a>
    <li><a href="#condor">How is BLCR different than "user-level" checkpointing libraries like Condor, etc.?</a>
    <li><a href="#platforms">What kinds of Linux systems does BLCR support?</a>
    <li><a href="#distros">What Linux distributions does BLCR work with?</a>
    <li><a href="#threads">Can BLCR checkpoint/restart multithreaded programs?</a>
    <li><a href="#procs">Can BLCR checkpoint/restart multi-process applications?</a>
    <li><a href="#limitations">Are there limits to the types of programs can BLCR checkpoint? </a>
    <li><a href="#callbacks">What if I want to checkpoint a program that uses resources that BLCR can't checkpoint?</a>
    <li><a href="#mpi">Does BLCR support checkpointing parallel/distributed applications?</a>
    <li><a href="#rootbuild">Do I need root access in order to use BLCR?</a>
    <li><a href="#batch">Is BLCR integrated with any batch systems, like PBS, Sun Grid Engine, Torque, LoadLeveler, etc? </a>
    <li><a href="#porting">How hard is it to port BLCR to an architecture that isn't currently supported?</a>
</ul>

<h3>Build/Install Questions</h3>
<ul>
    <li><a href="#patch">Does BLCR require a kernel patch?</a>
    <li><a href="#buildreqs">What do I need in order to build and use BLCR?</a>
    <li><a href="#unconfig">What if I my kernel sources are unconfigured?</a>
    <li><a href="#modvesion">How do I get past the error "linux/modversions.h:
    No such file or directory"?</a>
    <li><a href="#systemmap"> What if I get an error about my <tt>System.map</tt> file?</a>
    <li><a href="#kcompiler"> Why do I get "Invalid module format" when loading BLCR's kernel modules?</a>
</ul>

<h3>Usage Questions</h3>
<ul>
    <li><a href="#samepid">Why do I get this error: "Restart failed:
        Device or resource busy"?</a>
    <li><a href="#sig31">Why does my application dies with "Real-time signal 31"
        (or 32, etc.) when I try to checkpoint it?</a>
    <li><a href="#prelink">Why can I restart jobs on the original machine they ran
    on, but not a different node in my cluster?</a>
    <li><a href="#enoent">Why do I get the error "Restart failed: No such file or directory"?</a>
    <li><a href="#eperm">Why do I get the error "Restart failed: Permission denied"?</a>
    <li><a href="#preload">Why do I get the error "ERROR: ld.so: object 'libcr_run.so.0' from LD_PRELOAD cannot be preloaded: ignored"?</a>
    <li><a href="#nscd">Why do I get "vmadump: mmap failed: /var/db/nscd/[something]" in the system logs?</a>
    <li><a href="#hsperfdata">Why do I get "vmadump: mmap failed: /tmp/hsperfdata_[user]/[pid]" in the system logs?</a>
    <li><a href="#staticlink">Why can't I checkpoint my statically linked application?</a>
    <li><a href="#lam_cr_base_dir">When I checkpoint LAM MPI jobs, all the
    checkpoints wind up in my $HOME directory.  Can I use a different
    directory?</a>
</ul>

<h3>Additional Resources</h3>
<ul>
    <li><a href="#website">Where can I download BLCR?</a>
    <li><a href="#website">Where can I find more information?</a>
    <li><a href="#mailaddr">Is there a mailing list for BLCR?</a>
    <li><a href="#bugzilla">Where can I report BLCR bugs?</a>
</ul>

<hr>
<h3>General Questions</h3>

<h4><a name="whatisblcr">What is BLCR?</a></h4>
<blockquote>
    BLCR (Berkeley Lab Checkpoint/Restart) allows programs running on Linux to
    be "checkpointed" (written entirely to a file), and then later "restarted".
    BLCR can be found
    at <a href="http://ftg.lbl.gov/checkpoint">http://ftg.lbl.gov/checkpoint</a>.
</blockquote>

<h4><a name="sigstop">How is checkpoint/restart different than SIGSTOP/SIGCONT?</a></h4>
<blockquote>
      <p> Putting a process to sleep (via the SIGSTOP signal) implies stopping its
      execution.  Taking a checkpoint writes a snapshot of a process to disk:
      the process may either be allowed to continue running after the checkpoint is
      complete, or you can kill the process to release all of its resources .

      <p> With sleep, a process's resources are not all fully released (such as virtual
      memory, network connections, process id, etc.).  Checkpointing then killing a
      process fully releases all system resources.

      <p> Restarts from checkpoint files can be used across machine reboots, and/or even
      on different machines than the one that the checkpoint was taken on.  This is
      not true for SIGCONT.
</blockquote>

<h4><a name="condor">How is BLCR different than "user-level" checkpointing libraries like Condor, etc.?</a></h4>
<blockquote>
    BLCR performs checkpointing and restarting inside the linux kernel.  While
    this makes it less portable than solutions that use user-level libraries, it
    also means that it has full access to all kernel resources, and can thus
    restore resources (like process IDs) that user-level libraries cannot.
    This also allows BLCR to checkpoint/restart groups of processes (such as
    shell scripts and their subprocesses), together with the pipes that
    connect them.
</blockquote>

<h4><a name="platforms">What kinds of Linux systems does BLCR support?</a></h4>
<blockquote>
    BLCR runs on x86 and x86_64 (Opteron/EM64T) systems running Linux 2.6.x kernels.
    With the 0.8.2 release, we believe the following to work:
    <ul>
      <li>2.6.0 through 2.6.30 on x86 and x86_64.
    </ul>
    <p>BLCR 0.7.0 added experimental support for PPC (32-bit), and 0.6.0 added experimental
       support for PPC64 and ARM.  These three architectures have been tested as follows:
    <ul>
      <li>2.6.14 and newer on PPC
      <li>2.6.12 and newer on PPC64
      <li>2.6.17 and newer on ARM
    </ul>
    We are interested to hear of your success or failure with these three experimental
    architectures, especially on kernels older than those we have tested.
    <p>Note that 0.6.x was the last release series to support 2.4.x kernels.
</blockquote>

<h4><a name="distros">What Linux distributions does BLCR work with?</a></h4>
<blockquote>
    <p> BLCR uses a set of configure-based tests to determine which kernel features
    are available, and so in principle, BLCR should work with any 
    distribution that uses a supported CPU/kernel combination
    (see <a href="#platforms">above</a>).

    <p> Historically, BLCR has been tested with kernels from versions 9.x and 10.x of SuSE
    Linux; 
    CentOS 3.1; Fedora Core 2 through 8; and many vanilla Linux kernels (from kernel.org)
    from 2.6.0 on up.  We have not tested every single version of the kernel from every
    vendor, nor is each BLCR release retested against all distributions tested in the
    past.  However, we believe that BLCR should work on most distributions using
    kernels in the ranges given <a href="#platforms">above</a> (except where vendors
    may have applied patches that bring in problematic changes from kernels outside
    that range).

    <p> If after reading this question and the one <a href="#platforms">above</a> you
    believe your platform should be supported but cannot get BLCR to work, then please
    consult our <a href="#bugzilla">bug database</a> for possible solutions and then
    report the problem if you don't find it already reported.  We can't try every
    possible platform ourselves and count on user's bug reports to let us know when
    our testing has missed something.
</blockquote>

<h4><a name="threads">Can BLCR checkpoint/restart multithreaded programs?</a></h4>
<blockquote>
    Yes, BLCR can checkpoint both single- and multithreaded (pthreads) programs
    linked with the NPTL implementation of pthreads.
    <strong>Since with the 0.7.0 release, BLCR no longer provides full support
    for the older LinuxThreads implementation of pthreads.  Please contact us if
    you are able to devote some effort to restoring/maintaining such support in
    the future.</strong>
    <p> BLCR has not been tested with other threading packages,
    such as those used by some Java runtimes.  We are interested in hearing of
    both success and failure with other threading packages.
</blockquote>

<h4><a name="procs">Can BLCR checkpoint/restart multi-process applications?</a></h4>
<blockquote>
    Yes, starting with version 0.5.0 BLCR is able to save and restore groups of
    related processes together with the pipes that connect them.  To do this, BLCR
    must be given a single request that covers all the processes involved.  Currently
    there are three ways to specify a group request to BLCR:
    <ul>
      <li>A process "tree", consisting of a process and all its non-orphaned descendants.
         (An orphaned process is one who's parent has exited, leaving it as a child
         of the <tt>init</tt> process).
      <li>A POSIX process group, consisting of all processes with a given PGID (including
          orphaned processes, if any).
      <li>A POSIX session, consisting of all processes with a given SID (including
          orphaned processes, if any).
    </ul>

    <p>While BLCR can save and restore the pipes used for IPC among processes in these
    groups, it is unable at this time to deal with most other IPC mechanisms
    (see <a href="#limitations">next FAQ</a>).
</blockquote>

<h4><a name="limitations">Are there limits to the types of programs can BLCR checkpoint? </a></h4>
<blockquote>
    Yes. BLCR does not support checkpointing certain process resources. 
    While the following list is not exhaustive, it lists the most significant
    issues we are aware of.
<ul>
    <li>BLCR will not checkpoint and/or restore open sockets
    (TCP/IP, Unix domain, etc.).  At restart time any sockets will appear
    to have been closed.</li>

    <li>BLCR does not handle SysV IPC objects (man 5 ipc).  Such
    resources are silently ignored at checkpoint time and are not
    restored.</li>

    <li>If a checkpoint is taken of a process with any "zombie" children, then
    these children will not be recreated at restart time.  A "zombie" is
    defined as a process that has exited, but who's exit status has not yet
    been reaped by its parent (via <tt>wait()</tt> or a related function).
    This means that a wait()-family call made after a restart will never return
    a status for such a child.</li>
</ul>

    <p>If needed, applications can arrange to save any information necessary to
    recreate/reacquire such resources at restart time
    (see <a href="#callbacks">next FAQ</a>).
</blockquote>

<h4><a name="callbacks">What if I want to checkpoint a program that uses resources that BLCR can't checkpoint?</a></h4>
<blockquote>
    BLCR supports adding 'callbacks' to user-level code, which are called when a
    checkpoint is about to be performed, and when it is restarted (or continues
    on after the checkpoint).  This is how MPI communication can be handled
    (see <a href="#mpi">next FAQ</a>).

    <p>Full documentation of the callback interface is not yet documented because some of
    the interfaces are still subject to change.  However, the comments in the file
    <tt>libcr.h</tt> should provide enough to get started.
</blockquote>

<h4><a name="mpi">Does BLCR support checkpointing parallel/distributed applications?</a></h4>
<blockquote>
    Not by itself.  But by using checkpoint callbacks (see <a href="#callbacks">previous FAQ</a>).
    some MPI implementations, including LAM/MPI 7.x, MVAPICH2 0.9.8 and MPICH-V 1.0.x, have made themselves
    checkpointable by BLCR.  You can checkpoint/restart an MPI application running
    across an entire cluster of machines with BLCR, without any application code
    modifications, if you use one of these MPI implementations.

    <p>As of late November 2008, support in OpenMPI (the successor to LAM/MPI) is not available in
    a released version, but has been announced as a feature intended for the 1.3 release (and
    <em>is</em> available in the development branch).
</blockquote>

<h4><a name="rootbuild">Do I need root access in order to use BLCR?</a></h4>
<blockquote>
    Root access is needed to install the BLCR kernel modules.  However, once
    these are installed, any user can checkpoint and restart their own programs
    without needing root permission.
</blockquote>

<h4><a name="batch">Is BLCR integrated with any batch systems, like PBS, Sun Grid Engine, Torque, LoadLeveler, etc? </a></h4>
<blockquote>
    <p>Support for serial and parallel jobs is available in TORQUE, since in their 2.3 release.

    <p> Information on integration with SGE can be found
       <a href="http://gridengine.sunsource.net/howto/APSTC-TB-2004-005.pdf">here</a>.

    <p> Work is ongoing by third
    parties to integrate BLCR into other batch systems.  If you are
    interested in adding BLCR support to a job launcher/scheduler, please
    contact us!
</blockquote>

<h4><a name="porting">How hard is it to port BLCR to an architecture that isn't currently supported?</a></h4>
<blockquote>
    <p> Most of the architecture-specific code in BLCR is confined to small set of
    logic to save and restore the CPU-specific register set (vmadump) and some
    gcc inline assembly for atomic operations and special system calls.  The majority of
    BLCR's code base is entirely processor-independent.

    <p> If you are interested in
    seeing BLCR run on other chips, and are able to devote programmer resources,
    please contact us!  The Alpha platform is likely to be
    the easiest since vmadump already supports this architecture for
    Linux 2.6 kernels.
</blockquote>

<hr>
<h3>Build/Install Questions</h3>

<h4><a name="patch">Does BLCR require a kernel patch?</a></h4>
<blockquote>
    No. All of the kernel logic used by BLCR is implemented within kernel
    modules.  You can thus compile BLCR and load it into a running kernel (with
    <tt>modprobe</tt> or <tt>insmod</tt>) without needing to recompile your kernel or reboot.
</blockquote>

<h4><a name="buildreqs">What do I need in order to build and use BLCR?</a></h4>
<blockquote>
    A machine that is running a supported architecture (x86 and x86_64 are fully
supported and PPC/PPC64 and ARM are "Experimental") and
    Linux kernel 2.6.x.

    <p> A set of <b>configured</b> kernel headers that matches the kernel you wish to build against.
    By configured, we mean that <tt>include/linux/version.h</tt> and the files in
    <tt>include/linux/modules/</tt> match the target kernel.
    For many distributions a kernel-devel package is often
    enough if using the vendor's kernel.  For a custom kernel, the actual kernel build
    directory is often required.

    <p> The kernel's symbol table.  Normally the file <tt>/boot/System.map</tt>, or
    equivalent will serve this purpose.
</blockquote>

<h4><a name="unconfig">What if I my kernel sources are unconfigured?</a></h4>
<blockquote>
    BLCR needs to be able to examine a linux kernel source tree that has been
    configured, and this configuration must match the kernel that you will run
    BLCR against.  

    <p>If you do not have a configured linux kernel source tree, you may be able
    to create one fairly easily.  Many distributions provide a 'config' file
    that is all you need to easily produce a configured kernel source tree.
    Good places to look for a config file include
    <tt>/boot/config-2.6.5-1.358</tt> or <tt>/config-2.6.5-1.358</tt>.  In
    some distributions, the kernel is actually setup to include its
    configuration in <tt>/proc/config.gz</tt> (or <tt>/proc/config.bz2</tt>).
    If you can find any one of these files then we can proceed with the
    following steps:

<ol>
<li>Make a copy of the unconfigured source for the linux kernel you are using, and copy in the file you located:
<pre>
  $ cp -a /usr/src/kernel-source-2.6.5 /tmp/linux-2.6.5-1.358
  $ cd /tmp/linux-2.6.5-1.358
  $ cp [CONFIG_FILE] .config
</pre></li>
<li>Configure it using one of the following:
<ul>
<li>For kernels 2.6.6 and newer:
<pre>
  $ make modules_prepare
</pre></li>
<li>For 2.6.x kernels prior to 2.6.6:
<pre>
  $ make prepare-all scripts
</pre></li>
</ul>
<li>Once that is done, you should be able to configure BLCR using the newly
configured kernel source.  You should continue to use the <tt>System.map</tt>
file from your running kernel.  What you want is probably something like
<pre>
  $ ./configure --with-system-map=/boot/System.map-2.6.5-1.358 --with-linux=/tmp/linux-2.6.5-1.358.
</pre></li>
</ol>
</blockquote>

<h4><a name="modvesion">How do I get past the error "linux/modversions.h: No such file or directory"?</a></h4>
<blockquote>
    <p> Please try rebuilding blcr after commenting out the following six lines
    near the top of the files <tt>vmadump/vmadump.c</tt> and <tt>blcr_imports/imports.c</tt>:
<pre>
   #if defined(CONFIG_MODVERSIONS) &amp;&amp; ! defined(MODVERSIONS)
     #define MODVERSIONS
   #endif
   #if defined(MODVERSIONS)
     #include &lt;linux/modversions.h&gt;
   #endif
</pre>
    Let us know if your compilation still doesn't work.
</blockquote>

<h4><a name="systemmap">What if I get an error about my <tt>System.map</tt> file?</a></h4>
<blockquote>
    To build, BLCR needs to read the <tt>System.map</tt> file that corresponds to the
    kernel you will use BLCR with.  Generally, BLCR will find this file
    "automagically" during <tt>./configure</tt>, but some distributions do not
    provide it, and/or you may not keep yours in a standard place.

    <p>If you know where the correct <tt>System.map</tt> file is, use 
<pre>
  $ ./configure --with-system-map=PATH_TO_YOUR_SYSTEM.MAP
</pre>

    <p>If your <tt>System.map</tt> is absent, it may still be available as an optional RPM.
    For instance you may be able to get it by installing (depending
    on the release) either the <tt>kernel-source</tt> or <tt>kernel-devel</tt> RPMs
    for the kernel you will use BLCR with.

    <p>However, Fedora Core 2 and some of its derivatives are shipping a "stripped-down"
    <tt>System.map</tt> file.  If this is the case, BLCR will abort during configuration
    with an error stating
    that the <tt>System.map</tt> cannot be used.  You must install an additional RPM which
    contains a full <tt>System.map</tt> in order to build BLCR.  In Fedora Core 2
    the '<tt>kernel-debuginfo</tt>' RPM contains a full <tt>System.map</tt> file, which it
    will install into the <tt>/usr/lib/debug/boot</tt> directory.  BLCR's configure
    script will search this directory, but just to be certain you may still wish to pass
    '<tt>--with-system-map</tt>' to point configure at the correct <tt>System.map</tt> file.
  <blockquote>
    <strong>Important Note</strong>:  If you need to install the
    <tt>kernel-debuginfo</tt> RPM, <em> make sure the correct version
        is installed</em>.  Specifically, the 'arch' type must be the same.  If your
    kernel was built for the '<tt>i386</tt>' (or '<tt>i586</tt>', or
    '<tt>i686</tt>'), the kernel-debuginfo RPM must have the same value.  Thus,
    for an <tt>i586</tt> kernel, install
    '<tt>kernel-debuginfo-2.6.5-1.358.i586.rpm</tt>'.  To determine which
    kernel version you have, use
<pre>
  $ rpm -q kernel --qf '%{version}-%{release}.%{arch}\n'
</pre>
    To make sure that you have installed compatible <tt>kernel</tt> and
    <tt>kernel-debuginfo</tt> RPMs, use
<pre>
  $ rpm -q kernel kernel-debuginfo --qf '%{version}-%{release}.%{arch}\n'
</pre>
    (replace '<tt>kernel</tt>' with '<tt>kernel-smp</tt>' if you are using an SMP
    kernel).  You should see the same string, repeated twice.

    <p>If you try to use BLCR with the wrong <tt>System.map</tt>, BLCR will
    build without complaints, but will probably detect the problem when the
    <tt>blcr.o</tt> kernel module is loaded (it does this by comparing some
    well-known exported kernel symbols' addresses to those provided by the
    <tt>System.map</tt> file), and the module load will be aborted. 
  </blockquote>
</blockquote>

<h4><a name="kcompiler">Why do I get "Invalid module format" when loading BLCR's kernel modules?</a></h4>
<blockquote>
    Kernels from 2.6.0 through 2.6.18 check at module load time that the same compiler version (major and minor
    numbers) were used to build both the module and the kernel.  This is the error you will see
    if they don't match.  When this happens, you will need to reconfigure and build BLCR with
    the correct compiler.  When a module fails to load due to a version mismatch, you should
    be able to find a message in the system logs indicating the required compiler version:
<pre>
  blcr_imports: version magic '2.6.17 SMP mod_unload PENTIUM4 gcc-3.4' should be '2.6.17 SMP mod_unload PENTIUM4 gcc-3.2'
</pre>
    Alternatively, the following should find the "signature" in existing kernel modules:
<pre>
  $ find /lib/modules/`uname -r` -name '*.ko' -print | head -1 | xargs strings | grep gcc
  vermagic=2.6.17 SMP mod_unload PENTIUM4 gcc-3.2
</pre>
    In this case a gcc-3.2.X version is required.

    <p>Regardless of which method is used to find the correct version, you will need to
    reconfigure BLCR to use the correct compiler.  To do so, rerun <tt>configure</tt> with the
    addition of "KCC=/path/to/the/correct/gcc" to the command line to set the compiler used
    for building BLCR's kernel modules.
</blockquote>

<hr>
<h3>Usage Questions</h3>

<h4><a name="samepid">Why do I get this error: "Restart failed:
        Device or resource busy"?</a></h4>
<blockquote>
    This is because a resource needed into order to restart the process is already
    in use.  The most common problem is that another process already exists with the
    same pid (process ID)--the operating system will not allow you to create two
    programs with the same pid.   Very frequently this is because a user is trying
    to 'restart' a process from a checkpoint, when the original process they took
    the checkpoint of is still running!   

    <p>If you are unlucky enough that some other, unrelated process has grabbed the
    PID of your application, you must figure out some way to get rid of that
    process.  If you own the process, you can of course simply kill it (or
    checkpoint it!).  Otherwise, consider becoming root, or consulting your system
    administrator.  BLCR will not kill another process for you (this 'feature' would
    raise certain security issues).
</blockquote>

<h4><a name="prelink">Why can I restart jobs on the original machine they ran
    on, but not a different node in my cluster?</a></h4>
<blockquote>
    You should be able to restart a BLCR-checkpointed job on a different node
    (or set of nodes, for a parallel job), provided that all the nodes involved
    provide the exact same libraries and other files that your executable needs.
    By default BLCR does <em>not</em> save the contents of shared libraries that your program
    uses, nor does it save the contents of any files your program has
    <tt>open()</tt>'ed.  But so long as all of these libraries and/or files
    exist on another node, your program should restart fine.

    <p>Note that libraries must be <em>exactly</em> the same for a restart to
    work; if they are not the same size, for instance, restart will not work.
    If you've installed the same version of the operating system to all of your
    nodes (and you've updated them all the same way), you would think things would be fine.  
    However, some Linux distributions are using "prelinking", which is a
    method for assigning fixed addresses for shared libraries to load into
    executables.  Prelinking is a feature which enables applications that use many
    shared libraries to load faster.  The fixed address used by the same library
    on different nodes is often deliberately randomized (in order to defeat
    buffer overflow attacks that could otherwise rely on standard libraries
    being loaded at the same address on every machine with the same OS version).
    Alas, if the prelinked addresses are different, you will not be able to
    restart BLCR checkpoints on another node.

    <p> The solution for this problem is to disable prelinking on both the
    source and destination nodes of any process migration before starting any
    process you may wish to migrate.  For most cluster environments, that means
    disabling it on all nodes before using BLCR for migration.
    Prelinking is a systemwide setting, so you will need to be root.
    On Fedora Core 2, at least, the fix is to edit
    <tt>/etc/sysconfig/prelink</tt> and set '<tt>PRELINKING=no</tt>'.  The
    comments claim that this will cause prelinking to be undone automatically
    the next night.  We've never been patient, and instead "undo" prelinking
    immediately by running (as root)
<pre>  # /usr/sbin/prelink --undo --all</pre>
    Automating this process for an entire cluster depends on your specific environment.

    <p>Finally, be aware that BLCR 0.7.0 introduced the <tt>--save-*</tt> family
    of options to <tt>cr_checkpoint</tt> to cause the executable and/or shared
    libraries to be included in the context file.  This may significantly increase
    the size of the context files.  Therefore we recommend this approach only
    if you cannot ensure uniform library vesions (w/o prelinking) across the
    machines you wish to migrate among.
</blockquote>

<h4><a name="enoent">Why do I get the error "Restart failed: No such file or directory"?</a></h4>
<blockquote>
    This error normally means that a file that was open at the time the checkpoint
    was taken is no longer present (it is either completely gone, or perhaps just
    not present at the same pathname as it was previously).  You should examine your
    system logs (such as <tt>/var/log/messages</tt> or dmesg) for an indication of the
    file that caused the problem.  You will probably find a message like one of the following:
    <pre>    vmadump: mmap failed: /tmp/hsperfdata_[user]/[pid]<br>    Failed to open file '/tmp/foobar'<br></pre>
    In the case of files in a directory of the form  <tt>/tmp/hsperfdata_[user]</tt> see
    <a href="#hsperfdata">the hsdperfdata FAQ entry</a>.  For other files, there are a
    a couple of things you might try.
    <p> If the file is a temporary created by your application, it is
    possible that it has been removed when the application terminated.  For instance, if
    you checkpoint an application with the <tt>--term</tt> option to the <tt>cr_checkpoint</tt>
    utility, then <tt>SIGTERM</tt> was sent to the application, causing it to cleanup before
    terminating.  If this is the case, then passing <tt>--kill</tt> will cause the uncatchable
    signal <tt>SIGKILL</tt> to be sent, thus preventing any cleanups by the application.  Of
    course, if your application ran to completion and removed its temporary file at its
    normal conclusion, then you are on your own as to how to recover the file.
    <p> If you are trying to perform migration of a process from one machine to another, then
    it is possible that the file exists, but not at the full pathname that was saved by BLCR.
    This is especially true if network filesystem mountpoints differ between machines.  You
    may be able to work around such issues with symbolic links
    among directories, but BLCR provides a <tt>--relocate</tt> option to <tt>cr_restart</tt>
    that can easily deal with such situations.
    <p>Future versions of BLCR will make it possible to capture the contents of all open
    files, thus providing a mechanism to eliminate problems of this sort (at the cost of
    increased context file size, of course).
</blockquote>

<h4><a name="eperm">Why do I get the error "Restart failed: permission denied"?</a></h4>
<blockquote>
    There are multiple reasons why a restart would fail with this message, but the most
    common is filesystem permissions.  You should examine your system logs (such as
    <tt>/var/log/messages</tt> or dmesg) for an indication of the file that caused the
    problem.  You will probably find a message like one of the following:
    <pre>    vmadump: mmap failed: /var/db/nscd/hosts<br>    Failed to open file '/tmp/foobar'<br></pre>
    You should verify that our user has permission to access the file.  In the case
    of files in the directory <tt>/var/db/nscd</tt> see <a href="#nscd">the NSCD FAQ entry</a>.
</blockquote>

<h4><a name="preload">
Why do I get the error "ERROR: ld.so: object 'libcr_run.so.0' from LD_PRELOAD cannot be preloaded: ignored"?
</a></h4>
<blockquote>
    This error is almost always a symptom that BLCR's shared libraries
    are not located for one of two reasons:
    <ul>
    <li> The libraries have been installed in a "standard" directory, but
    the utility <tt>ldconfig</tt> has not been run as root to update the
    loader cache.
    <li> The libraries have been installed in a "non-standard" directory
    and the directory has not been added to the LD_LIBRARY_PATH environment
    variable.
    </ul> 
    Please see, respectively, the sections "Updating ld.so.cache" and
    "Configuring Users' environments" in the BLCR Admin Guide for
    information on resolving these installation issues.
</blockquote>

<h4><a name="nscd">Why do I get "vmadump: mmap failed: /var/db/nscd/[something]" in the system logs?</a></h4>
<blockquote>
    The files in the directory <tt>/var/db/nscd/</tt> are created by the Name Service
    Cache Daemon (NSCD, for short) and are protected against normal file accesses.
    When the NSCD is enabled, certain C library operations talk to the daemon which uses
    filedescriptor passing to allow the application to mmap these files.  Unfortunately, BLCR
    cannot replay the filedescriptor passing (there is no way to know that a given filedescriptor
    was obtained in this way).  This leaves BLCR to rely on normal filesystem permissions when
    trying to reestablish the <tt>mmap()</tt>, which in this case finds insufficient permissions.
    <p>
    C library functions that may use NSCD include host database lookups (such as
    <tt>gethostbyname()</tt>), user database lookups (such as <tt>getpwuid()</tt>) and
    other database lookup functions.
    <p>
    There is no BLCR-level work-around for this type of failure, and we don't know of
    any way to disable use of NSCD on a per-process basis.  Therefore, we recommend that
    if you encounter this problem, you should disable (or preferable just uninstall)
    NSCD on your system.  You should consult the documentation for your specific
    Linux distribution for instructions.
</blockquote>

<h4><a name="hsperfdata">Why do I get "vmadump: mmap failed: /tmp/hsperfdata_[user]/[pid]" in the system logs?</a></h4>
<blockquote>
    Directories of the form <tt>/tmp/hsperfdata_[user]/</tt> are created by Sun's Java
    runtime environment (JRE), and the individual files in this directory are removed
    when the application exits.  This includes exits due to fatal signals, so if one
    checkpoints a Java application with the <tt>--term</tt> option to the <tt>cr_checkpoint</tt>
    utility, then <tt>SIGTERM</tt> was sent to the application, causing it to cleanup its
    <tt>hsperfdata</tt> file before terminating.  In this case, passing <tt>--kill</tt> will
    cause the uncatchable signal <tt>SIGKILL</tt> to be sent, thus preventing any cleanups by
    the JRE.
    <p> If you are trying to perform migration of a process from one machine to another, then
    you may try manually copying the file from one machine to another.  While this is not a
    safe practice in general, we have had reports of success from users who have tried this.
    <p> If you are running Matlab but don't need the Java features, you can avoid this issue
    entirely by passing <tt>--nojre</tt> when launching Matlab.
</blockquote>

<h4><a name="sig31">Why does my application dies with "Real-time signal 31" (or 32, etc.) when I try to checkpoint it?</a></h4>
<blockquote>
    This error was possible in older releases of BLCR when an application was not
    checkpointable.  This should not happen in release 0.5.0 or newer and should
    be reported as a bug if seen.
</blockquote>

<h4><a name="staticlink">Why can't I checkpoint my statically linked application?</a></h4>
<blockquote>
  If you can checkpoint and restart a dynamically linked application correctly, but
  cannot do so with the same application linked statically, this FAQ entry is for you.
  There are multiple reasons why BLCR may have problems with statically executables.
  <ul>
    <li>The <tt>cr_run</tt> utility only supports dynamic executables<br>
    If you wish to checkpoint an unmodified executable, the typical recipe is
<pre>
  $ cr_run my_app my_args
</pre>
    However, the <tt>cr_run</tt> utility does its work using the "LD_PRELOAD"
    environment variable to force loading of BLCR's support code into the
    address space the applications.  That mechanism is only functional for
    dynamically linked executables.  There is no magic we can perform today
    that will resolve this (though in the future we'd like to replace our
    use of LD_PRELOAD with a kernel-side mechanism).  So, you'll need to
    relink any statically linked executables to include BLCR support.
    </li>
    <li>Linking BLCR's libraries statically takes special care<br>
    OK, we've told you why <tt>cr_run</tt> doesn't work and told you to
    relink.  You tried linking with <tt>-lcr_run</tt> and/or <tt>-lcr</tt>
    and still can't get a checkpoint to work.  What went wrong?<p>
    Please reread the "Cautionary linker notes" section of the BLCR Users
    Guide.  You need a <tt>-u</tt> option in addition the the <tt>-l</tt>
    or the static linking will simply ignore BLCR's library.
    </li>
    <li>BLCR doesn't support LinuxThreads<br>
    Ok, what else could go wrong?  You've followed the guidance given
    in the "Cautionary linker notes" section of the BLCR Users Guide
    when you linked your application.  You even ran
<pre>
  $ nm my_app | grep link_me
</pre>
    to be sure the symbol you specified with <tt>-u</tt> is linked in.
    However, you are seeing weird crashes of your application when you
    try to checkpoint.<p>
    The culprit might be LinuxThreads.  Why?  Because at the time this
    FAQ entry is being written, there are many Linux distributions
    that install the static libs for LinuxThreads in the default library
    search path, and with the NPTL static libs elsewhere.  The resolution
    could be as simple as linking your application with <tt>-L/usr/lib/nptl</tt>
    or <tt>-L/usr/lib64/nptl</tt>, perhaps by setting an "LDFLAGS" variable
    (though it is possible that your distribution has picked some other location).<p>
    While it is not strictly required due to binary compatibility between
    LinuxThreads and NPTL, we'd recommend that you at least consider a
    recompile with <tt>-I/usr/include/nptl</tt> in CFLAGS.<p>
    Note, of course, that if BLCR's utilities are statically linked to
    LinuxThreads, then they need to be rebuilt too.  See the BLCR
    Admin Guide for more information on that.
    </li>
  </ul>
</blockquote>

<h4><a name="lam_cr_base_dir">When I checkpoint LAM MPI jobs, all the
        checkpoints wind up in my $HOME directory.  Can I use a different
        directory?</a></h4>
<blockquote>
    By default, LAM/MPI will use $HOME as the location for storing the
    checkpoint files for all the processes involved in an MPI job, unless it was
    configured at build time with '<tt>configure --with-cr-file-dir=/somewhere/else</tt>', 
    in which case '<tt>/somewhere/else</tt>' will be the default location.  So
    rebuilding LAM is one (rather slow and painful) way to change where
    checkpoints are stored.

    <p>A much easier solution is to set the LAM '<tt>cr_base_dir</tt>' SSI
    parameter for each individual job that you wish to have use a different
    directory for storing checkpoints.  This can either be done by setting the
    '<tt>$LAM_MPI_SSI_cr_base_dir</tt>' environment variable to the full path of
    the directory you want to use, or by setting the '<tt>cr_base_dir</tt>'
    parameter on the command line:
<pre>
  $ mpirun -np 2 -ssi cr_base_dir /somewhere/else a.out
</pre>
    See the <a href="http://www.lam-mpi.org/using/docs/">LAM Documentation</a>
    for more details, especially the "Available MPI Modules | Checkpoint/Restart
    of MPI Jobs" section in the User's Guide.
</blockquote>

<hr>
<h3>Additional Resources</h3>

<h4><a name="website">Where can I download BLCR?<br>
Where can I find more information?</a></h4>
<blockquote>
    To download the BLCR software, or for links to all the available information
    about BLCR, please
    visit <a href="http://ftg.lbl.gov/checkpoint">http://ftg.lbl.gov/checkpoint</a>.
</blockquote>

<h4><a name="mailaddr">Is there a mailing list for BLCR?</a></h4>
<blockquote>
    There is a mailing list of BLCR developers and some of the users
    at <a href="mailto:checkpoint@lbl.gov">checkpoint@lbl.gov</a> and
    which is
    archived <a href="http://www.nersc.gov/hypermail/checkpoint/">here</a>.

    <p> This list is managed my majordomo.  So, to subscribe just e-mail
    "<tt>subscribe checkpoint</tt>" (in the message <em>body</em> and without the quotes)
    to <a href="mailto:majordomo@lbl.gov">majordomo@lbl.gov</a>.
</blockquote>

<h4><a name="bugzilla">Where can I report BLCR bugs?</a></h4>
<blockquote>
    If you think you've found a bug in BLCR, please <em>do</em> let us know about it.
    There are many kernel-dependencies in BLCR and we could easily have missed testing on
    a system like yours.  We count on user's bug reports to help ensure wide testing
    coverage.

    <p> The BLCR bug database is managed by a Bugzilla, located
    at <a href="http://mantis.lbl.gov/bugzilla">http://mantis.lbl.gov/bugzilla</a>.

    <p> Before reporting a bug, you are encouraged to search the database to see if
    a bug report exists for your problem.  For some issues a solution can be
    found in just a day.  So, a patch to fix your problem may already be attached
    to an existing bug report.  BLCR is just one of a group of projects managed on
    this server, so be sure to select product "BLCR" in your queries.
</blockquote>

</body>
</html>
blcr-dkms 0.8.2-15ubuntu2 / usr / src / blcr-0.8.2 / doc / html / FAQ.html
