Re: [PATCH] generic: add a less thorough testing mode for fsync-err program

[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]



On Thu, Jun 29, 2017 at 09:29:47AM -0400, jlayton@xxxxxxxxxx wrote:
> From: Jeff Layton <jlayton@xxxxxxxxxx>
> 
> Currently we just have this test run on a whitelist of filesystems, but
> it would be best to be able to run it on all of them. The problem is
> that a lot of filesystems basically shut down once they hit metadata
> errors.
> 
> Allow the fsync-err testcase to operate in two different modes. One
> mode just does basic testing to ensure that we get an error back on
> all fd's when we fsync. The other does a more thorough test to ensure
> that we get back 0 on subsequent fsyncs when there hasn't been any
> write activity.
> 
> For now, we just opt-in to the more thorough testing on certain
> filesystems: xfs, ext3 and ext4 on the generic test. All other
> filesystems will run in simple mode.
> 
> Signed-off-by: Jeff Layton <jlayton@xxxxxxxxxx>
> ---
>  common/rc         |  9 +++++++++
>  src/fsync-err.c   | 51 +++++++++++++++++++++++++++++++++------------------
>  tests/generic/441 | 22 +++++++++++++++-------
>  3 files changed, 57 insertions(+), 25 deletions(-)
> 
> diff --git a/common/rc b/common/rc
> index 2972f89e9527..83675364cf24 100644
> --- a/common/rc
> +++ b/common/rc
> @@ -1738,6 +1738,15 @@ _require_test()
>      touch ${RESULT_DIR}/require_test
>  }
>  
> +_has_logdev()
> +{
> +	ret=0
> +	[ -z "$SCRATCH_LOGDEV" -o ! -b "$SCRATCH_LOGDEV" ] &&
> +	[ "$USE_EXTERNAL" != yes ] && ret=1

This doesn't seem right. _has_logdev = ($SCRATCH_LOGDEV is block
device) && ($USE_EXTERNAL set to yes), so we should set ret=1 if any of
these conditions is not met. But this code returns 0 (true) if I only
have SCRATCH_LOGDEV set but not USE_EXTERNAL.

And once we have _has_logdev, I think we can refactor _require_logdev to
use _has_logdev too.

> +
> +	return $ret
> +}
> +
>  # this test needs a logdev
>  #
>  _require_logdev()
> diff --git a/src/fsync-err.c b/src/fsync-err.c
> index 5b3bdd3ada07..4b0205cf2fd4 100644
> --- a/src/fsync-err.c
> +++ b/src/fsync-err.c
> @@ -13,6 +13,7 @@
>  #include <string.h>
>  #include <unistd.h>
>  #include <getopt.h>
> +#include <stdbool.h>
>  
>  /*
>   * btrfs has a fixed stripewidth of 64k, so we need to write enough data to
> @@ -25,7 +26,7 @@
>  
>  static void usage()
>  {
> -	printf("Usage: fsync-err [ -b bufsize ] [ -n num_fds ] -d dmerror path <filename>\n");
> +	printf("Usage: fsync-err [ -b bufsize ] [ -n num_fds ] [ -s ] -d dmerror path <filename>\n");
>  }
>  
>  int main(int argc, char **argv)
> @@ -35,8 +36,9 @@ int main(int argc, char **argv)
>  	char *dmerror_path = NULL;
>  	char *cmdbuf;
>  	size_t cmdsize, bufsize = DEFAULT_BUFSIZE;
> +	bool simple_mode = false;
>  
> -	while ((i = getopt(argc, argv, "b:d:n:")) != -1) {
> +	while ((i = getopt(argc, argv, "b:d:n:s")) != -1) {
>  		switch (i) {
>  		case 'b':
>  			bufsize = strtol(optarg, &buf, 0);
> @@ -55,6 +57,15 @@ int main(int argc, char **argv)
>  				return 1;
>  			}
>  			break;
> +		case 's':
> +			/*
> +			 * Many filesystems will continue to throw errors after
> +			 * fsync has already advanced to the current error,
> +			 * due to metadata writeback failures or other
> +			 * issues. Allow those fs' to opt out of more thorough
> +			 * testing.
> +			 */
> +			simple_mode = true;
>  		}
>  	}
>  
> @@ -154,16 +165,18 @@ int main(int argc, char **argv)
>  		}
>  	}
>  
> -	for (i = 0; i < numfds; ++i) {
> -		ret = fsync(fd[i]);
> -		if (ret < 0) {
> -			/*
> -			 * We did a failed write and fsync on each fd before.
> -			 * Now the error should be clear since we've not done
> -			 * any writes since then.
> -			 */
> -			printf("Third fsync on fd[%d] failed: %m\n", i);
> -			return 1;
> +	if (!simple_mode) {
> +		for (i = 0; i < numfds; ++i) {
> +			ret = fsync(fd[i]);
> +			if (ret < 0) {
> +				/*
> +				 * We did a failed write and fsync on each fd before.
> +				 * Now the error should be clear since we've not done
> +				 * any writes since then.
> +				 */
> +				printf("Third fsync on fd[%d] failed: %m\n", i);
> +				return 1;
> +			}
>  		}
>  	}
>  
> @@ -185,12 +198,14 @@ int main(int argc, char **argv)
>  		return 1;
>  	}
>  
> -	for (i = 0; i < numfds; ++i) {
> -		ret = fsync(fd[i]);
> -		if (ret < 0) {
> -			/* The error should still be clear */
> -			printf("fsync after healing device on fd[%d] failed: %m\n", i);
> -			return 1;
> +	if (!simple_mode) {
> +		for (i = 0; i < numfds; ++i) {
> +			ret = fsync(fd[i]);
> +			if (ret < 0) {
> +				/* The error should still be clear */
> +				printf("fsync after healing device on fd[%d] failed: %m\n", i);
> +				return 1;
> +			}
>  		}
>  	}
>  
> diff --git a/tests/generic/441 b/tests/generic/441
> index 2215b64db9a7..e8244224e097 100755
> --- a/tests/generic/441
> +++ b/tests/generic/441
> @@ -45,15 +45,23 @@ _cleanup()
>  . ./common/dmerror
>  
>  # real QA test starts here
> -_supported_fs ext2 ext3 ext4 xfs
>  _supported_os Linux
>  _require_scratch
>  
> -# Generally, we want to avoid journal errors in this test. Ensure that
> -# journalled fs' have a logdev.
> -if [ "$FSTYP" != "ext2" ]; then
> -	_require_logdev
> -fi
> +# Generally, we want to avoid journal errors on the extended testcase. Only
> +# set the -r flag if we have a logdev
> +sflag='-s'

This part seems confusing, comments said "set the -r flag" but '-s' is
actually used.

> +case $FSTYP in
> +	btrfs)
> +		_notrun "btrfs has a specialized test for this"
> +		;;
> +	ext3|ext4|xfs)
> +		# Do the more thorough test if we have a logdev
> +		_has_logdev && sflag=''
> +		;;
> +	*)
> +		;;
> +esac
>  
>  _require_dm_target error
>  _require_test_program fsync-err
> @@ -70,7 +78,7 @@ _require_fs_space $SCRATCH_MNT 65536
>  
>  testfile=$SCRATCH_MNT/fsync-err-test
>  
> -$here/src/fsync-err -d $here/src/dmerror $testfile
> +$here/src/fsync-err $sflag -d $here/src/dmerror $testfile

Better to append this command to $seqres.full too for debug purpose, so
we know if we're running with or without '-s' option.

Thanks,
Eryu

>  
>  # success, all done
>  _dmerror_load_working_table
> -- 
> 2.13.0
> 
--
To unsubscribe from this list: send the line "unsubscribe fstests" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html



[Index of Archives]     [Linux Filesystems Development]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux