Looks like with force_fsync NuKeeper doesn't work properly on sandbox HDDs.

01396_inactive_replica_cleanup_nodes_zookeeper - need fix test
01149_zookeeper_mutation_stuck_after_replace_partition - something strange happened, test is probably flaky
01079_parallel_alter_add_drop_column_zookeeper - Timeout - DDLWorker hung due to All connection tries failed while connecting to ZooKeeper
01018_ddl_dictionaries_concurrent_requrests - database test_ha0hh4 was not created due to Coordination::Exception: Operation timeout, path: /clickhouse
00992_system_parts_race_condition_zookeeper_long - Coordination::Exception: Operation timeout, path: /test/clickhouse/db/default/log/query-0000000032
00965_set_index_string_functions - database test_1jv5ht was not created - Coordination::Exception: Operation timeout, path: /clickhouse
00732_quorum_insert_lost_part_zookeeper - Replica doesn't have part 20181115_0_0_0 which was successfully written to quorum of other replicas. Send query to another replica or disable 'select_sequential_consistency' setting - seems like part was written to another replica

Functional stateless tests (release, DatabaseReplicated) — fail: 0, passed: 2781, skipped: 105
Functional stateful tests (release, DatabaseReplicated) — fail: 0, passed: 108, skipped: 3

00652_replicated_mutations_default_database_zookeeper - fails because for_subquery table is not replicated, need add to skip list
00620_optimize_on_nonleader_replica_zookeeper - need SYSTEM SYNC REPLICA rename1 before the first select to make it work with additional replicas (or just add it to skip list)

Functional stateless tests (release, DatabaseReplicated) — fail: 0, passed: 2784, skipped: 107
Functional stateful tests (release, DatabaseReplicated) — fail: 0, passed: 108, skipped: 3

Need to merge with #22274.

Functional stateful tests (release, DatabaseReplicated) — fail: 0, passed: 109, skipped: 3
Functional stateless tests (release, DatabaseReplicated) — fail: 0, passed: 2812, skipped: 107

Waiting for tests.

Need to fix flaky check?

Stress thread -- OOM

2021.03.30 12:41:50.266427 [ 301 ] {} <Fatal> Application: Child process was terminated by signal 9 (KILL). If it is not done by 'forcestop' command or manually, the possible cause is OOM Killer (see 'dmesg' and look at the '/var/log/kern.log' for the details).