Handle and detect #13709's unlock regression (#14161)

In #13709, as in #11294 before it, it turns out that 63a26454 still had
the same failure mode as when it was first landed as d1d47691, and
fails to unlock certain datasets that formerly worked.

Rather than reverting it again, let's add handling to just throw out
the accounting metadata that failed to unlock when that happens, as
well as a test with a pre-broken pool image to ensure that we never get
bitten by this again.

Fixes: #13709

Signed-off-by: Rich Ercolani <[email protected]>
Reviewed-by: Brian Behlendorf <[email protected]>
Reviewed-by: Tony Hutter <[email protected]>

4 files changed, 48 insertions, 1 deletions

diff --git a/tests/runfiles/common.run b/tests/runfiles/common.run
index 323c37a3d..73ca69993 100644
--- a/tests/runfiles/common.run
+++ b/tests/runfiles/common.run
@@ -911,7 +911,7 @@ tests = [
     'userquota_007_pos', 'userquota_008_pos', 'userquota_009_pos',
     'userquota_010_pos', 'userquota_011_pos', 'userquota_012_neg',
     'userspace_001_pos', 'userspace_002_pos', 'userspace_encrypted',
-    'userspace_send_encrypted']
+    'userspace_send_encrypted', 'userspace_encrypted_13709']
 tags = ['functional', 'userquota']
 
 [tests/functional/vdev_zaps]
diff --git a/tests/zfs-tests/tests/Makefile.am b/tests/zfs-tests/tests/Makefile.am
index 5b8458b73..39eb44f73 100644
--- a/tests/zfs-tests/tests/Makefile.am
+++ b/tests/zfs-tests/tests/Makefile.am
@@ -368,6 +368,7 @@ nobase_dist_datadir_zfs_tests_tests_DATA += \
 	functional/upgrade/upgrade_common.kshlib \
 	functional/user_namespace/user_namespace.cfg \
 	functional/user_namespace/user_namespace_common.kshlib \
+	functional/userquota/13709_reproducer.bz2 \
 	functional/userquota/userquota.cfg \
 	functional/userquota/userquota_common.kshlib \
 	functional/vdev_zaps/vdev_zaps.kshlib \
@@ -1935,6 +1936,7 @@ nobase_dist_datadir_zfs_tests_tests_SCRIPTS += \
 	functional/userquota/userspace_003_pos.ksh \
 	functional/userquota/userspace_encrypted.ksh \
 	functional/userquota/userspace_send_encrypted.ksh \
+	functional/userquota/userspace_encrypted_13709.ksh \
 	functional/vdev_zaps/cleanup.ksh \
 	functional/vdev_zaps/setup.ksh \
 	functional/vdev_zaps/vdev_zaps_001_pos.ksh \
diff --git a/tests/zfs-tests/tests/functional/userquota/13709_reproducer.bz2 b/tests/zfs-tests/tests/functional/userquota/13709_reproducer.bz2
new file mode 100644
index 000000000..9c3168279
--- /dev/null
+++ b/tests/zfs-tests/tests/functional/userquota/13709_reproducer.bz2
diff --git a/tests/zfs-tests/tests/functional/userquota/userspace_encrypted_13709.ksh b/tests/zfs-tests/tests/functional/userquota/userspace_encrypted_13709.ksh
new file mode 100755
index 000000000..9c1d84775
--- /dev/null
+++ b/tests/zfs-tests/tests/functional/userquota/userspace_encrypted_13709.ksh
@@ -0,0 +1,45 @@
+#!/bin/ksh -p
+#
+# This file and its contents are supplied under the terms of the
+# Common Development and Distribution License ("CDDL"), version 1.0.
+# You may only use this file in accordance with the terms of version
+# 1.0 of the CDDL.
+#
+# A full copy of the text of the CDDL should have accompanied this
+# source.  A copy of the CDDL is also available via the Internet at
+# http://www.illumos.org/license/CDDL.
+#
+
+. $STF_SUITE/include/libtest.shlib
+. $STF_SUITE/tests/functional/userquota/userquota_common.kshlib
+
+#
+# DESCRIPTION:
+# Avoid allowing #11294/#13709 to recur a third time.
+#
+# So we hardcode a copy of a pool with this bug, try unlocking it,
+# and fail on error. Simple.
+
+function cleanup
+{
+	destroy_pool $POOLNAME
+	rm -f $FILEDEV
+}
+
+log_onexit cleanup
+
+FILEDEV="$TEST_BASE_DIR/userspace_13709"
+POOLNAME="testpool_13709"
+
+log_assert "ZFS should be able to unlock pools with #13709's failure mode"
+
+log_must bzcat $STF_SUITE/tests/functional/userquota/13709_reproducer.bz2 > $FILEDEV
+
+log_must zpool import -d $FILEDEV $POOLNAME
+
+echo -e 'password\npassword\n' | log_must zfs mount -al
+
+# Cleanup
+cleanup
+
+log_pass "#13709 not happening here"