diff options
author | Don Brady <[email protected]> | 2019-06-22 16:41:21 -0700 |
---|---|---|
committer | Brian Behlendorf <[email protected]> | 2019-06-22 16:41:21 -0700 |
commit | a9cd8bfde73a78a0ba02e25b712fe28d11019191 (patch) | |
tree | adec993bdaa74a2cc43ce5624395efef737b7b0d | |
parent | fb6e6f1ffb2aa18711c8c3c825c1090bd97561f0 (diff) |
Let zfs mount all tolerate in-progress mounts
The zfs-mount service can unexpectedly fail to start when zfs
encounters a mount that is in progress. This service uses
zfs mount -a, which has a window between the time it checks if
the dataset was mounted and when the actual mount (via mount.zfs
binary) occurs.
The reason for the racing mounts is that both zfs-mount.target
and zfs-share.target are allowed to execute concurrently after
the import. This is more of an issue with the relatively recent
addition of parallel mounting, and we should consider serializing
the mount and share targets.
Reviewed-by: Brian Behlendorf <[email protected]>
Reviewed by: John Kennedy <[email protected]>
Reviewed-by: Allan Jude <[email protected]>
Signed-off-by: Don Brady <[email protected]>
Closes #8881
-rw-r--r-- | cmd/zfs/zfs_main.c | 19 |
1 files changed, 18 insertions, 1 deletions
diff --git a/cmd/zfs/zfs_main.c b/cmd/zfs/zfs_main.c index 9357bd2f3..634303f80 100644 --- a/cmd/zfs/zfs_main.c +++ b/cmd/zfs/zfs_main.c @@ -6580,8 +6580,25 @@ share_mount_one(zfs_handle_t *zhp, int op, int flags, char *protocol, return (1); } - if (zfs_mount(zhp, options, flags) != 0) + if (zfs_mount(zhp, options, flags) != 0) { + /* + * Check if a mount sneaked in after we checked + */ + if (!explicit && + libzfs_errno(g_zfs) == EZFS_MOUNTFAILED) { + usleep(10 * MILLISEC); + libzfs_mnttab_cache(g_zfs, B_FALSE); + + if (zfs_is_mounted(zhp, NULL)) { + (void) fprintf(stderr, gettext( + "Ignoring previous 'already " + "mounted' error for '%s'\n"), + zfs_get_name(zhp)); + return (0); + } + } return (1); + } break; } |