On 9/30/20 10:59 AM, Darrick J. Wong wrote:
> On Wed, Sep 30, 2020 at 10:54:42AM -0500, Eric Sandeen wrote:
>> Today, if one tries to repair a filesystem with a realtime subvol but
>> forgets to specify the rtdev on the command line, the result sounds dire:
>>
>> Phase 1 - find and verify superblock...
>> xfs_repair: filesystem has a realtime subvolume
>> xfs_repair: realtime device init failed
>> xfs_repair: cannot repair this filesystem. Sorry.
>>
>> We can be a bit more helpful, following the log device example:
>>
>> Phase 1 - find and verify superblock...
>> This filesystem has a realtime subvolume. Specify rt device with the -r option.
>>
>> Signed-off-by: Eric Sandeen <sandeen@xxxxxxxxxx>
>> ---
>>
>> diff --git a/libxfs/init.c b/libxfs/init.c
>> index cb8967bc..65cc3d4c 100644
>> --- a/libxfs/init.c
>> +++ b/libxfs/init.c
>> @@ -429,9 +429,9 @@ rtmount_init(
>> if (sbp->sb_rblocks == 0)
>> return 0;
>> if (mp->m_rtdev_targp->dev == 0 && !(flags & LIBXFS_MOUNT_DEBUGGER)) {
>> - fprintf(stderr, _("%s: filesystem has a realtime subvolume\n"),
>> - progname);
>> - return -1;
>> + fprintf(stderr, _("This filesystem has a realtime subvolume. "
>> + "Specify rt device with the -r option.\n"));
>
> _("%s: realtime subvolume must be specified.\n")?
except if we were called by xfs_copy, it can't handle it at all.
I'll need to move this to repair-specific code.
> --D
>
>> + exit(1);
>> }
>> mp->m_rsumlevels = sbp->sb_rextslog + 1;
>> mp->m_rsumsize =
>>
>
