Fix 'zpool add' handling of nested interior VDEVs

When replacing a faulted device which was previously handled by a spare
multiple levels of nested interior VDEVs will be present in the pool
configuration; the following example illustrates one of the possible
situations:

   NAME                          STATE     READ WRITE CKSUM
   testpool                      DEGRADED     0     0     0
     raidz1-0                    DEGRADED     0     0     0
       spare-0                   DEGRADED     0     0     0
         replacing-0             DEGRADED     0     0     0
           /var/tmp/fault-dev    UNAVAIL      0     0     0  cannot open
           /var/tmp/replace-dev  ONLINE       0     0     0
         /var/tmp/spare-dev1     ONLINE       0     0     0
       /var/tmp/safe-dev         ONLINE       0     0     0
   spares
     /var/tmp/spare-dev1         INUSE     currently in use

This is safe and allowed, but get_replication() needs to handle this
situation gracefully to let zpool add new devices to the pool.

Reviewed-by: George Melikov <mail@gmelikov.ru>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: loli10K <ezomori.nozomu@gmail.com>
Closes #6678 
Closes #6996

cmd/zpool/zpool_vdev.c

@@ -861,9 +861,11 @@ get_replication(nvlist_t *nvroot, boolean_t fatal)
 
 				/*
 				 * If this is a replacing or spare vdev, then
-				 * get the real first child of the vdev.
+				 * get the real first child of the vdev: do this
+				 * in a loop because replacing and spare vdevs
+				 * can be nested.
 				 */
-				if (strcmp(childtype,
+				while (strcmp(childtype,
 				    VDEV_TYPE_REPLACING) == 0 ||
 				    strcmp(childtype, VDEV_TYPE_SPARE) == 0) {
 					nvlist_t **rchild;

tests/runfiles/linux.run

@@ -257,7 +257,7 @@ tests = ['zpool_add_001_pos', 'zpool_add_002_pos', 'zpool_add_003_pos',
     'zpool_add_004_pos', 'zpool_add_005_pos', 'zpool_add_006_pos',
     'zpool_add_007_neg', 'zpool_add_008_neg', 'zpool_add_009_neg',
     'zpool_add_010_pos',
-    'add-o_ashift', 'add_prop_ashift']
+    'add-o_ashift', 'add_prop_ashift', 'add_nested_replacing_spare']
 tags = ['functional', 'cli_root', 'zpool_add']
 
 [tests/functional/cli_root/zpool_attach]

tests/zfs-tests/tests/functional/cli_root/zpool_add/Makefile.am

@@ -15,4 +15,5 @@ dist_pkgdata_SCRIPTS = \
 	zpool_add_009_neg.ksh \
 	zpool_add_010_pos.ksh \
 	add-o_ashift.ksh \
-	add_prop_ashift.ksh
+	add_prop_ashift.ksh \
+	add_nested_replacing_spare.ksh

tests/zfs-tests/tests/functional/cli_root/zpool_add/add_nested_replacing_spare.ksh

@@ -0,0 +1,111 @@
+#!/bin/ksh -p
+#
+# CDDL HEADER START
+#
+# The contents of this file are subject to the terms of the
+# Common Development and Distribution License (the "License").
+# You may not use this file except in compliance with the License.
+#
+# You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
+# or http://www.opensolaris.org/os/licensing.
+# See the License for the specific language governing permissions
+# and limitations under the License.
+#
+# When distributing Covered Code, include this CDDL HEADER in each
+# file and include the License file at usr/src/OPENSOLARIS.LICENSE.
+# If applicable, add the following below this CDDL HEADER, with the
+# fields enclosed by brackets "[]" replaced with your own identifying
+# information: Portions Copyright [yyyy] [name of copyright owner]
+#
+# CDDL HEADER END
+#
+
+#
+# Copyright 2017, loli10K <ezomori.nozomu@gmail.com>. All rights reserved.
+#
+
+. $STF_SUITE/include/libtest.shlib
+. $STF_SUITE/tests/functional/cli_root/zpool_create/zpool_create.shlib
+
+#
+# DESCRIPTION:
+#	'zpool add' works with nested replacing/spare vdevs
+#
+# STRATEGY:
+#	1. Create a redundant pool with a spare device
+#	2. Manually fault a device, wait for the hot-spare and then replace it:
+#	   this creates a situation where replacing and spare vdevs are nested.
+#	3. Verify 'zpool add' is able to add new devices to the pool.
+#
+
+verify_runnable "global"
+
+function cleanup
+{
+	zed_stop
+	zed_cleanup
+	log_must zinject -c all
+	destroy_pool $TESTPOOL
+	log_must rm -f $DATA_DEVS $SPARE_DEVS
+}
+
+log_assert "'zpool add' works with nested replacing/spare vdevs"
+log_onexit cleanup
+
+FAULT_DEV="$TEST_BASE_DIR/fault-dev"
+SAFE_DEV1="$TEST_BASE_DIR/safe-dev1"
+SAFE_DEV2="$TEST_BASE_DIR/safe-dev2"
+SAFE_DEV3="$TEST_BASE_DIR/safe-dev3"
+SAFE_DEVS="$SAFE_DEV1 $SAFE_DEV2 $SAFE_DEV3"
+REPLACE_DEV="$TEST_BASE_DIR/replace-dev"
+ADD_DEV="$TEST_BASE_DIR/add-dev"
+DATA_DEVS="$FAULT_DEV $SAFE_DEVS $REPLACE_DEV $ADD_DEV"
+SPARE_DEV1="$TEST_BASE_DIR/spare-dev1"
+SPARE_DEV2="$TEST_BASE_DIR/spare-dev2"
+SPARE_DEVS="$SPARE_DEV1 $SPARE_DEV2"
+
+# We need ZED running to work with spares
+zed_setup
+zed_start
+# Clear events from previous runs
+zed_events_drain
+
+for type in "mirror" "raidz1" "raidz2" "raidz3"
+do
+	# 1. Create a redundant pool with a spare device
+	truncate -s $SPA_MINDEVSIZE $DATA_DEVS $SPARE_DEVS
+	log_must zpool create $TESTPOOL $type $FAULT_DEV $SAFE_DEVS
+	log_must zpool add $TESTPOOL spare $SPARE_DEV1
+
+	# 2.1 Fault a device, verify the spare is kicked in
+	log_must zinject -d $FAULT_DEV -e nxio -T all -f 100 $TESTPOOL
+	log_must zpool scrub $TESTPOOL
+	log_must wait_vdev_state $TESTPOOL $FAULT_DEV "UNAVAIL" 60
+	log_must wait_vdev_state $TESTPOOL $SPARE_DEV1 "ONLINE" 60
+	log_must wait_hotspare_state $TESTPOOL $SPARE_DEV1 "INUSE"
+	log_must check_state $TESTPOOL "" "DEGRADED"
+
+	# 2.2 Replace the faulted device: this creates a replacing vdev inside a
+	#     spare vdev
+	log_must zpool replace $TESTPOOL $FAULT_DEV $REPLACE_DEV
+	log_must wait_vdev_state $TESTPOOL $REPLACE_DEV "ONLINE" 60
+	zpool status | awk -v poolname="$TESTPOOL" -v type="$type" 'BEGIN {s=""}
+	    $1 ~ poolname {c=4}; (c && c--) { s=s$1":" }
+	    END { if (s != poolname":"type"-0:spare-0:replacing-0:") exit 1; }'
+	if [[ $? -ne 0 ]]; then
+		log_fail "Pool does not contain nested replacing/spare vdevs"
+	fi
+
+	# 3. Verify 'zpool add' is able to add new devices
+	log_must zpool add $TESTPOOL spare $SPARE_DEV2
+	log_must wait_hotspare_state $TESTPOOL $SPARE_DEV2 "AVAIL"
+	log_must zpool add -f $TESTPOOL $ADD_DEV
+	log_must wait_vdev_state $TESTPOOL $ADD_DEV "ONLINE" 60
+
+	# Cleanup
+	log_must zinject -c all
+	destroy_pool $TESTPOOL
+	log_must rm -f $DATA_DEVS $SPARE_DEVS
+done
+
+log_pass "'zpool add' works with nested replacing/spare vdevs"