[Cluster-devel] [PATCH] gfs2_lockcapture: The script now returns a correct exit code when the script exits.

[Steven Whitehouse <swhiteho@redhat.com>]

Hi,

Looks ok to me. ACK,

Steve.

On Thu, 2012-12-20 at 10:39 -0500, sbradley at redhat.com wrote:
> From: Shane Bradley <sbradley@redhat.com>
> 
> If there was files(not directories) created then the script will exit with 0,
> else 1 since there was no lockdump files captured on any run of capturing the
> data. Updated examples in man page. Change the default output path so that the
> node name is not in the path. In addition I change how the clusternode name is
> parsed.
> 
> Signed-off-by: Shane Bradley <sbradley@redhat.com>
> ---
>  gfs2/scripts/gfs2_lockcapture | 49 ++++++++++++++++++++++++++++---------------
>  1 file changed, 32 insertions(+), 17 deletions(-)
> 
> diff --git a/gfs2/scripts/gfs2_lockcapture b/gfs2/scripts/gfs2_lockcapture
> index 1a64188..2b3421c 100644
> --- a/gfs2/scripts/gfs2_lockcapture
> +++ b/gfs2/scripts/gfs2_lockcapture
> @@ -33,7 +33,7 @@ import tarfile
>  sure only 1 instance of this script is running at any time.
>  @type PATH_TO_PID_FILENAME: String
>  """
> -VERSION_NUMBER = "0.9-2"
> +VERSION_NUMBER = "0.9-3"
>  MAIN_LOGGER_NAME = "%s" %(os.path.basename(sys.argv[0]))
>  PATH_TO_DEBUG_DIR="/sys/kernel/debug"
>  PATH_TO_PID_FILENAME = "/var/run/%s.pid" %(os.path.basename(sys.argv[0]))
> @@ -570,11 +570,10 @@ def getClusterNode(listOfGFS2Names):
>              stdout = runCommandOutput("corosync-quorumtool", ["-l"])
>              if (not stdout == None):
>                  for line in stdout.split("\n"):
> -                    splitLine = line.split()
> -                    if (len(splitLine) == 4):
> -                        if (splitLine[0].strip().rstrip() == thisNodeID):
> -                            clusternodeName = splitLine[3]
> -                            break;
> +                    if (line.find("local") >=0):
> +                        splitLine = line.split(" (local)")
> +                        clusternodeName = splitLine[0].split()[2]
> +                        break;
>      # If a clusternode name and cluster name was found then return a new object
>      # since this means this cluster is part of cluster.
>      if ((len(clusterName) > 0) and (len(clusternodeName) > 0)):
> @@ -815,6 +814,10 @@ def gatherGFS2LockDumps(pathToDSTDir, listOfGFS2Filesystems):
>      and filesystem name for each item in the list. For example:
>      "f18cluster:mygfs2vol1"
>  
> +    @return: Returns True if files(not directories) were copied to the
> +    destination directory.
> +    @rtype: Boolean
> +
>      @param pathToDSTDir: This is the path to directory where the files will be
>      copied to.
>      @type pathToDSTDir: String
> @@ -825,13 +828,19 @@ def gatherGFS2LockDumps(pathToDSTDir, listOfGFS2Filesystems):
>      lockDumpType = "gfs2"
>      pathToSrcDir = os.path.join(PATH_TO_DEBUG_DIR, lockDumpType)
>      pathToOutputDir = os.path.join(pathToDSTDir, lockDumpType)
> +    # The number of files that were copied
> +    fileCopiedCount = 0
>      for dirName in os.listdir(pathToSrcDir):
>          pathToCurrentDir = os.path.join(pathToSrcDir, dirName)
>          if ((os.path.isdir(pathToCurrentDir)) and (dirName in listOfGFS2Filesystems)):
>              message = "Copying the lockdump data for the %s filesystem: %s" %(lockDumpType.upper(), dirName)
>              logging.getLogger(MAIN_LOGGER_NAME).debug(message)
> -            copyDirectory(pathToCurrentDir, pathToOutputDir)
> -
> +            copySuccessful = copyDirectory(pathToCurrentDir, pathToOutputDir)
> +            if (copySuccessful and os.path.exists(os.path.join(pathToOutputDir, dirName))):
> +                fileCopiedCount = len(os.listdir(os.path.join(pathToOutputDir, dirName)))
> +    # If the number of files(not directories) copied was greater than zero then files were copied
> +    # succesfully.
> +    return (fileCopiedCount > 0)
>  # ##############################################################################
>  # Get user selected options
>  # ##############################################################################
> @@ -936,18 +945,18 @@ class OptionParserExtended(OptionParser):
>          self.print_version()
>          examplesMessage = "\n"
>          examplesMessage = "\nPrints information about the available GFS2 filesystems that can have lockdump data captured."
> -        examplesMessage += "\n$ %s -i\n" %(self.__commandName)
> +        examplesMessage += "\n# %s -i\n" %(self.__commandName)
>  
>          examplesMessage += "\nIt will do 3 runs of gathering the lockdump information in 10 second intervals for only the"
>          examplesMessage += "\nGFS2 filesystems with the names myGFS2vol2,myGFS2vol1. Then it will archive and compress"
>          examplesMessage += "\nthe data collected. All of the lockdump data will be written to the directory: "
>          examplesMessage += "\n/tmp/2012-11-12_095556-gfs2_lockcapture and all the questions will be answered with yes.\n"
> -        examplesMessage += "\n$ %s -r 3 -s 10 -t -n myGFS2vol2,myGFS2vol1 -o /tmp/2012-11-12_095556-gfs2_lockcapture -y\n" %(self.__commandName)
> +        examplesMessage += "\n# %s -r 3 -s 10 -t -n myGFS2vol2,myGFS2vol1 -o /tmp/2012-11-12_095556-gfs2_lockcapture -y\n" %(self.__commandName)
>  
>          examplesMessage += "\nIt will do 2 runs of gathering the lockdump information in 25 second intervals for all the"
>          examplesMessage += "\nmounted GFS2 filesystems. Then it will archive and compress the data collected. All of the"
>          examplesMessage += "\nlockdump data will be written to the directory: /tmp/2012-11-12_095556-gfs2_lockcapture.\n"
> -        examplesMessage += "\n$ %s -r 2 -s 25 -t -o /tmp/2012-11-12_095556-gfs2_lockcapture\n" %(self.__commandName)
> +        examplesMessage += "\n# %s -r 2 -s 25 -t -o /tmp/2012-11-12_095556-gfs2_lockcapture\n" %(self.__commandName)
>          OptionParser.print_help(self)
>          print examplesMessage
>  
> @@ -1001,7 +1010,8 @@ class ExtendOption (Option):
>  # ###############################################################################
>  if __name__ == "__main__":
>      """
> -    When the script is executed then this code is ran.
> +    When the script is executed then this code is ran. If there was files(not
> +    directories) created then 0 will be returned, else a 1 is returned.
>      """
>      try:
>          # #######################################################################
> @@ -1106,7 +1116,7 @@ if __name__ == "__main__":
>          # #######################################################################
>          pathToOutputDir = cmdLineOpts.pathToOutputDir
>          if (not len(pathToOutputDir) > 0):
> -            pathToOutputDir = "%s" %(os.path.join("/tmp", "%s-%s-%s" %(time.strftime("%Y-%m-%d_%H%M%S"), clusternode.getClusterNodeName(), os.path.basename(sys.argv[0]))))
> +            pathToOutputDir = "%s" %(os.path.join("/tmp", "%s-%s" %(time.strftime("%Y-%m-%d_%H%M%S"), os.path.basename(sys.argv[0]))))
>          # #######################################################################
>          # Backup any existing directory with same name as current output
>          # directory.
> @@ -1139,8 +1149,12 @@ if __name__ == "__main__":
>          # Gather data and the lockdumps.
>          # #######################################################################
>          if (cmdLineOpts.numberOfRuns <= 0):
> -            message = "The number of runs should be greater than zero."
> +            message = "The number of runs must be greater than zero."
> +            logging.getLogger(MAIN_LOGGER_NAME).warning(message)
>              exitScript(errorCode=1)
> +        # If GFS2 lockdump files were successfully copied to output directory
> +        # then the exit code will be set to 0, else the exit code will be 1.
> +        exitCode = 1
>          for i in range(1,(cmdLineOpts.numberOfRuns + 1)):
>              # The current log count that will start at 1 and not zero to make it
>              # make sense in logs.
> @@ -1189,14 +1203,15 @@ if __name__ == "__main__":
>              lockDumpType = "gfs2"
>              message = "Pass (%d/%d): Gathering the %s lock dumps for the host." %(i, cmdLineOpts.numberOfRuns, lockDumpType.upper())
>              logging.getLogger(MAIN_LOGGER_NAME).debug(message)
> -            gatherGFS2LockDumps(pathToOutputRunDir, clusternode.getMountedGFS2FilesystemNames())
> +            if(gatherGFS2LockDumps(pathToOutputRunDir, clusternode.getMountedGFS2FilesystemNames())):
> +                exitCode = 0
>              # Gather log files
>              message = "Pass (%d/%d): Gathering the log files for the host." %(i, cmdLineOpts.numberOfRuns)
>              logging.getLogger(MAIN_LOGGER_NAME).debug(message)
>              gatherLogs(os.path.join(pathToOutputRunDir, "logs"))
>              # Sleep between each run if secondsToSleep is greater than or equal
>              # to 0 and current run is not the last run.
> -            if ((cmdLineOpts.secondsToSleep >= 0) and (i <= (cmdLineOpts.numberOfRuns))):
> +            if ((cmdLineOpts.secondsToSleep >= 0) and (i < (cmdLineOpts.numberOfRuns))):
>                  message = "The script will sleep for %d seconds between each run of capturing the lockdump data." %(cmdLineOpts.secondsToSleep)
>                  logging.getLogger(MAIN_LOGGER_NAME).info(message)
>                  time.sleep(cmdLineOpts.secondsToSleep)
> @@ -1228,4 +1243,4 @@ if __name__ == "__main__":
>      # #######################################################################
>      # Exit the application with zero exit code since we cleanly exited.
>      # #######################################################################
> -    exitScript()
> +    exitScript(errorCode=exitCode)