[PATCH] btrfs-progs: for mixed group check opt before default raid profile is enforced

[PATCH] btrfs-progs: for mixed group check opt before default raid profile is enforced

[Anand Jain]

This fixes the regression introduced with the patch

    btrfs-progs: avoid write to the disk before sure to create fs

what happened with this patch is it missed the check to see if the
user has the option set before pushing the defaults.

Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
 mkfs.c |   21 +++++++++++----------
 1 files changed, 11 insertions(+), 10 deletions(-)

diff --git a/mkfs.c b/mkfs.c
index ec717be..2156150 100644
--- a/mkfs.c
+++ b/mkfs.c
@@ -1359,17 +1359,13 @@ int main(int ac, char **av)
 	if (is_vol_small(file)) {
 		printf("SMALL VOLUME: forcing mixed metadata/data groups\n");
 		mixed = 1;
-		if (metadata_profile != data_profile) {
-			if (metadata_profile_opt || data_profile_opt) {
-				fprintf(stderr,
-	"With mixed block groups data and metadata profiles must be the same\n");
-				exit(1);
-			}
-		}
 	}
+
 	/*
 	* Set default profiles according to number of added devices.
-	* For mixed groups defaults are single/single.
+	* For mixed groups defaults are single/single
+	* however if metadata_profile_opt || data_profile_opt is set
+	* metadata_profile and data_profile must be same
 	*/
 	if (!mixed) {
 		if (!metadata_profile_opt) {
@@ -1387,8 +1383,13 @@ int main(int ac, char **av)
 				BTRFS_BLOCK_GROUP_RAID0 : 0; /* raid0 or single */
 		}
 	} else {
-		metadata_profile = 0;
-		data_profile = 0;
+		if (metadata_profile_opt || data_profile_opt) {
+			if (metadata_profile != data_profile) {
+				fprintf(stderr,
+	"With mixed block groups data and metadata profiles must be the same\n");
+				exit(1);
+			}
+		}
 	}
 
 	ret = test_num_disk_vs_raid(metadata_profile, data_profile,
-- 
1.7.1

Re: [PATCH] btrfs-progs: for mixed group check opt before default raid profile is enforced

[Goffredo Baroncelli]

Hi Anand,

On 2013-11-15 05:34, Anand Jain wrote:
> This fixes the regression introduced with the patch
> 
>     btrfs-progs: avoid write to the disk before sure to create fs
> 
> what happened with this patch is it missed the check to see if the
> user has the option set before pushing the defaults.
> 
> Signed-off-by: Anand Jain <anand.jain@oracle.com>
> ---
>  mkfs.c |   21 +++++++++++----------
>  1 files changed, 11 insertions(+), 10 deletions(-)
> 
> diff --git a/mkfs.c b/mkfs.c
> index ec717be..2156150 100644
> --- a/mkfs.c
> +++ b/mkfs.c
> @@ -1359,17 +1359,13 @@ int main(int ac, char **av)
>  	if (is_vol_small(file)) {
>  		printf("SMALL VOLUME: forcing mixed metadata/data groups\n");
>  		mixed = 1;
> -		if (metadata_profile != data_profile) {
> -			if (metadata_profile_opt || data_profile_opt) {
> -				fprintf(stderr,
> -	"With mixed block groups data and metadata profiles must be the same\n");
> -				exit(1);
> -			}
> -		}
>  	}
> +
>  	/*
>  	* Set default profiles according to number of added devices.
> -	* For mixed groups defaults are single/single.
> +	* For mixed groups defaults are single/single
> +	* however if metadata_profile_opt || data_profile_opt is set
> +	* metadata_profile and data_profile must be same
>  	*/
>  	if (!mixed) {
>  		if (!metadata_profile_opt) {
> @@ -1387,8 +1383,13 @@ int main(int ac, char **av)
>  				BTRFS_BLOCK_GROUP_RAID0 : 0; /* raid0 or single */
>  		}
>  	} else {
> -		metadata_profile = 0;
> -		data_profile = 0;
> +		if (metadata_profile_opt || data_profile_opt) {
> +			if (metadata_profile != data_profile) {
> +				fprintf(stderr,
> +	"With mixed block groups data and metadata profiles must be the same\n");

Please prefix the error message with "ERROR: " otherwise is not so easy to 
understand that this is an error message and that the filesystem was not created.

> +				exit(1);
> +			}
> +		}
>  	}
>  
>  	ret = test_num_disk_vs_raid(metadata_profile, data_profile,
> 


It works for me (see below), however I suggest to prefix the error message 
"With mixed block group..." with the "ERROR" string, otherwise it is not so obvious that 
something went wrong.

-------

# test: make a raid1 filesystem in a small volume to force the "mixed" mode
# expected results: the volume must have a Data+Metadata, RAID1 chunk
# result: test PASSED

ghigo@venice:/tmp$ sudo ~/btrfs/btrfs-progs/mkfs.btrfs -K -m raid1 -d raid1 /dev/loop[01]
SMALL VOLUME: forcing mixed metadata/data groups

WARNING! - Btrfs v0.20-rc1-595-g9f0c53f IS EXPERIMENTAL
WARNING! - see http://btrfs.wiki.kernel.org before using

Turning ON incompat feature 'mixed-bg': mixed data and metadata block groups
Turning ON incompat feature 'extref': increased hardlink limit per file to 65536
Created a data/metadata chunk of size 8388608
adding device /dev/loop1 id 2
fs created label (null) on /dev/loop0
	nodesize 4096 leafsize 4096 sectorsize 4096 size 202.00MiB
Btrfs v0.20-rc1-595-g9f0c53f
ghigo@venice:/tmp$ sudo mount /dev/loop0 /mnt/test
ghigo@venice:/tmp$ btrfs fi df /mnt/test/
System, RAID1: total=8.00MB, used=4.00KB
System: total=4.00MB, used=0.00
Data+Metadata, RAID1: total=64.00MB, used=28.00KB
Data+Metadata: total=8.00MB, used=0.00
ghigo@venice:/tmp$ sudo umount /mnt/test




# test: make a filesystem in a small volume to force the "mixed" mode 
#  with different profiles for data and metadata
# expected results: mkfs must fails with an error message
# result: test PASSED

ghigo@venice:/tmp$ sudo ~/btrfs/btrfs-progs/mkfs.btrfs -K -m dup -d raid1 /dev/loop[01]
SMALL VOLUME: forcing mixed metadata/data groups
With mixed block groups data and metadata profiles must be the same

BR
G.Baroncelli
-- 
gpg @keyserver.linux.it: Goffredo Baroncelli (kreijackATinwind.it>
Key fingerprint BBF5 1610 0B64 DAC6 5F7D  17B2 0EDA 9B37 8B82 E0B5

[PATCH v2] btrfs-progs: for mixed group check opt before default raid profile is enforced

[Anand Jain]

This fixes the regression introduced with the patch

    btrfs-progs: avoid write to the disk before sure to create fs

what happened with this patch is it missed the check to see if the
user has the option set before pushing the defaults.

Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
 v2: accepts G.Baroncelli sugested
     rebased on integration-20131114

 mkfs.c |   18 +++++++++---------
 1 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/mkfs.c b/mkfs.c
index 4e99c16..de1beed 100644
--- a/mkfs.c
+++ b/mkfs.c
@@ -1375,14 +1375,8 @@ int main(int ac, char **av)
 	if (is_vol_small(file)) {
 		printf("SMALL VOLUME: forcing mixed metadata/data groups\n");
 		mixed = 1;
-		if (metadata_profile != data_profile) {
-			if (metadata_profile_opt || data_profile_opt) {
-				fprintf(stderr,
-	"With mixed block groups data and metadata profiles must be the same\n");
-				exit(1);
-			}
-		}
 	}
+
 	/*
 	* Set default profiles according to number of added devices.
 	* For mixed groups defaults are single/single.
@@ -1404,8 +1398,14 @@ int main(int ac, char **av)
 		}
 	} else {
 		u32 best_leafsize = max_t(u32, sysconf(_SC_PAGESIZE), sectorsize);
-		metadata_profile = 0;
-		data_profile = 0;
+
+		if (metadata_profile_opt || data_profile_opt) {
+			if (metadata_profile != data_profile) {
+				fprintf(stderr,
+	"ERROR: With mixed block groups data and metadata profiles must be the same\n");
+				exit(1);
+			}
+		}
 
 		if (!leaf_forced) {
 			leafsize = best_leafsize;
-- 
1.7.1

Re: [PATCH v2] btrfs-progs: for mixed group check opt before default raid profile is enforced

[Chris Mason]

Quoting Anand Jain (2013-11-15 06:11:09)
> This fixes the regression introduced with the patch
> 
>     btrfs-progs: avoid write to the disk before sure to create fs
> 
> what happened with this patch is it missed the check to see if the
> user has the option set before pushing the defaults.

Thanks Goffredo for tracking it down, and Anand for the quick fix.

-chris