NAME

       error::buildid - build-id verification failures

DESCRIPTION

       Because  systemtap's script translation / execution stages may be executed at different times and places,
       it is sometimes necessary to verify certain invariants.  One such invariant  is  that  if  a  script  was
       informed  by  translate-time  analysis of executables, then those same executables need to be used at run
       time.   This  checking  is  done  based  upon  the  build-id,  a  binary  hash  that  modern  (post-2007)
       compilers/toolchains add as an NT_GNU_BUILD_ID ELF note to object files and executables.  Use the readelf
       -n command to examine the build-ids of binaries, if you are interested.

       Only  scripts  are sensitive to executables' build-ids: generally those that perform deep analysis of the
       binaries or their debuginfo.  For example, scripts that place .function  or  .statement  probes,  or  use
       stack  backtrace-related tapset functions may be sensitive.  Other scripts that rely only on process.mark
       or  kernel.trace  probes  do  not  require  debuginfo.   See  the  DWARF   DEBUGINFO   section   in   the
       stapprobes(3stap) man page.

       During  translation, systemtap saves a copy of the relevant files' build-ids within the compiled modules.
       At run-time, the modules compare the saved ones to the actual run-time build-ids in  memory.   The  error
       message  indicates  that they did not match, so the module will decline placing a probe that was computed
       based upon obsolete data.  This is important for safety, as placing  them  at  an  inappropriate  address
       could  crash the programs.  However, this is not necessarily a fatal error, since probes unrelated to the
       mismatching binaries may operate.

       A build-id mismatch could be caused by a few different situations.  The main one is where the  executable
       versions  or  architecture  were  different between the systemtap translation and execution times/places.
       For example, one may run a stap-server on a slightly different  version  of  the  OS  distribution.   The
       kernel running on the workstation may be slightly different from the version being targeted - perhaps due
       to  a  pending  kernel  upgrade  leaving  different  files  on disk versus running in memory.  If your OS
       distribution uses separate debuginfo packages, the split .IR .debug files may not exactly match the  main
       binaries.

SEE ALSO

       http://fedoraproject.org/wiki/Releases/FeatureBuildId,             stap(1),            stapprobes(3stap),
       warning::debuginfo(7stap), error::reporting(7stap)

                                                                                           ERROR::BUILDID(7stap)