datastore/tests/docker_scripts/stress_runner.sh

#!/bin/bash
# shellcheck disable=SC2094
# shellcheck disable=SC2086
# shellcheck disable=SC2024

set -ex

# Avoid overlaps with previous runs
dmesg --clear

ln -s /repo/tests/datastore-test /usr/bin/datastore-test

# shellcheck source=../stateless/stress_tests.lib
source /repo/tests/docker_scripts/stress_tests.lib


install_packages package_folder

# Thread Fuzzer allows to check more permutations of possible thread scheduling
# and find more potential issues.
export THREAD_FUZZER_CPU_TIME_PERIOD_US=1000
export THREAD_FUZZER_SLEEP_PROBABILITY=0.01
export THREAD_FUZZER_SLEEP_TIME_US_MAX=100000

export THREAD_FUZZER_pthread_mutex_lock_BEFORE_MIGRATE_PROBABILITY=1
export THREAD_FUZZER_pthread_mutex_lock_AFTER_MIGRATE_PROBABILITY=1
export THREAD_FUZZER_pthread_mutex_unlock_BEFORE_MIGRATE_PROBABILITY=1
export THREAD_FUZZER_pthread_mutex_unlock_AFTER_MIGRATE_PROBABILITY=1

export THREAD_FUZZER_pthread_mutex_lock_BEFORE_SLEEP_PROBABILITY=0.001
export THREAD_FUZZER_pthread_mutex_lock_AFTER_SLEEP_PROBABILITY=0.001
export THREAD_FUZZER_pthread_mutex_unlock_BEFORE_SLEEP_PROBABILITY=0.001
export THREAD_FUZZER_pthread_mutex_unlock_AFTER_SLEEP_PROBABILITY=0.001
export THREAD_FUZZER_pthread_mutex_lock_BEFORE_SLEEP_TIME_US_MAX=10000

export THREAD_FUZZER_pthread_mutex_lock_AFTER_SLEEP_TIME_US_MAX=10000
export THREAD_FUZZER_pthread_mutex_unlock_BEFORE_SLEEP_TIME_US_MAX=10000
export THREAD_FUZZER_pthread_mutex_unlock_AFTER_SLEEP_TIME_US_MAX=10000

export THREAD_FUZZER_EXPLICIT_SLEEP_PROBABILITY=0.001
export THREAD_FUZZER_EXPLICIT_MEMORY_EXCEPTION_PROBABILITY=0.001

export USE_ENCRYPTED_STORAGE=$((RANDOM % 2))

export ZOOKEEPER_FAULT_INJECTION=1
# Initial run without S3 to create system.*_log on local file system to make it
# available for dump via clickhouse-local
configure

# run before start_minio to have valid aws creds
cd /repo && python3 /repo/ci/jobs/scripts/datastore_proc.py logs_export_config || echo "ERROR: Failed to create log export config"

cd /repo && python3 /repo/ci/jobs/scripts/datastore_proc.py start_minio stateless || { echo "Failed to start minio"; exit 1; }
cd /repo && python3 /repo/ci/jobs/scripts/datastore_proc.py start_azurite || { echo "Failed to start azurite"; exit 1; }

# Start Redpanda (Kafka-compatible broker) so that Kafka engine tests work and
# do not leave behind broken StorageKafka tables whose background threads cause
# the server to freeze under sanitizers during the post-stress restart. Fail fast
# if the broker cannot be started: continuing without it would reintroduce the
# very failure mode this mitigation is here to prevent.
bash /repo/ci/jobs/scripts/functional_tests/setup_kafka.sh || { echo "Failed to start Kafka (Redpanda)"; exit 1; }

start_server || { echo "Failed to start server"; exit 1; }

cd /repo && python3 /repo/ci/jobs/scripts/datastore_proc.py logs_export_start || echo "ERROR: Failed to start log exports"

clickhouse-client --query "CREATE DATABASE datasets"
clickhouse-client < /repo/tests/docker_scripts/create.sql
bash /repo/tests/docker_scripts/create_tpcds.sh
bash /repo/tests/docker_scripts/create_tpch.sh
clickhouse-client --query "SHOW TABLES FROM datasets"
clickhouse-client --query "SHOW TABLES FROM tpcds"
clickhouse-client --query "SHOW TABLES FROM tpch"

clickhouse-client --query "CREATE DATABASE IF NOT EXISTS test"

stop_server
mv /var/log/clickhouse-server/clickhouse-server.log /var/log/clickhouse-server/clickhouse-server.initial.log

# Randomize cache policies.
cache_policy=""
if [ $((RANDOM % 2)) -eq 1 ]; then
    cache_policy="SLRU"
else
    cache_policy="LRU"
fi

echo "Using cache policy: $cache_policy"

if [ "$cache_policy" = "SLRU" ]; then
    sed -i.tmp "s|<cache_policy>LRU</cache_policy>|<cache_policy>SLRU</cache_policy>|" /etc/clickhouse-server/config.d/storage_conf*.xml
fi

start_server || { echo "Failed to start server"; exit 1; }

clickhouse-client --query "SYSTEM STOP THREAD FUZZER"

clickhouse-client --query "SHOW TABLES FROM datasets"
clickhouse-client --query "SHOW TABLES FROM tpcds"
clickhouse-client --query "SHOW TABLES FROM tpch"
clickhouse-client --query "SHOW TABLES FROM test"

if [[ "$USE_S3_STORAGE_FOR_MERGE_TREE" == "1" ]]; then
    TEMP_POLICY="s3_cache"
elif [[ "$USE_AZURE_STORAGE_FOR_MERGE_TREE" == "1" ]]; then
    TEMP_POLICY="azure_cache"
else
    random=$((RANDOM % 3))
    if [[ $random -eq 0 ]]; then
        TEMP_POLICY="default"
        echo "Using local storage policy"
    elif [[ $random -eq 1 ]]; then
        TEMP_POLICY="s3_cache"
        export USE_S3_STORAGE_FOR_MERGE_TREE=1
        echo "Using s3 storage policy"
    elif [[ $random -eq 2 ]]; then
        TEMP_POLICY="azure_cache"
        export USE_AZURE_STORAGE_FOR_MERGE_TREE=1
        echo "Using azure storage policy"
    fi
fi


clickhouse-client --query "CREATE TABLE test.hits_s3 (WatchID UInt64,  JavaEnable UInt8,  Title String,  GoodEvent Int16,
    EventTime DateTime,  EventDate Date,  CounterID UInt32,  ClientIP UInt32,  ClientIP6 FixedString(16),  RegionID UInt32,
    UserID UInt64,  CounterClass Int8,  OS UInt8,  UserAgent UInt8,  URL String,  Referer String,  URLDomain String,  RefererDomain String,
    Refresh UInt8,  IsRobot UInt8,  RefererCategories Array(UInt16),  URLCategories Array(UInt16), URLRegions Array(UInt32),
    RefererRegions Array(UInt32),  ResolutionWidth UInt16,  ResolutionHeight UInt16,  ResolutionDepth UInt8,  FlashMajor UInt8,
    FlashMinor UInt8,  FlashMinor2 String,  NetMajor UInt8,  NetMinor UInt8, UserAgentMajor UInt16,  UserAgentMinor FixedString(2),
    CookieEnable UInt8, JavascriptEnable UInt8,  IsMobile UInt8,  MobilePhone UInt8,  MobilePhoneModel String,  Params String,
    IPNetworkID UInt32,  TraficSourceID Int8, SearchEngineID UInt16,  SearchPhrase String,  AdvEngineID UInt8,  IsArtifical UInt8,
    WindowClientWidth UInt16,  WindowClientHeight UInt16,  ClientTimeZone Int16,  ClientEventTime DateTime,  SilverlightVersion1 UInt8,
    SilverlightVersion2 UInt8,  SilverlightVersion3 UInt32,  SilverlightVersion4 UInt16,  PageCharset String,  CodeVersion UInt32,
    IsLink UInt8,  IsDownload UInt8,  IsNotBounce UInt8,  FUniqID UInt64,  HID UInt32,  IsOldCounter UInt8, IsEvent UInt8,
    IsParameter UInt8,  DontCountHits UInt8,  WithHash UInt8, HitColor FixedString(1),  UTCEventTime DateTime,  Age UInt8,
    Sex UInt8,  Income UInt8,  Interests UInt16,  Robotness UInt8,  GeneralInterests Array(UInt16), RemoteIP UInt32,
    RemoteIP6 FixedString(16),  WindowName Int32,  OpenerName Int32,  HistoryLength Int16,  BrowserLanguage FixedString(2),
    BrowserCountry FixedString(2),  SocialNetwork String,  SocialAction String,  HTTPError UInt16, SendTiming Int32,
    DNSTiming Int32,  ConnectTiming Int32,  ResponseStartTiming Int32,  ResponseEndTiming Int32,  FetchTiming Int32,
    RedirectTiming Int32, DOMInteractiveTiming Int32,  DOMContentLoadedTiming Int32,  DOMCompleteTiming Int32,
    LoadEventStartTiming Int32,  LoadEventEndTiming Int32, NSToDOMContentLoadedTiming Int32,  FirstPaintTiming Int32,
    RedirectCount Int8, SocialSourceNetworkID UInt8,  SocialSourcePage String,  ParamPrice Int64, ParamOrderID String,
    ParamCurrency FixedString(3),  ParamCurrencyID UInt16, GoalsReached Array(UInt32),  OpenstatServiceName String,
    OpenstatCampaignID String,  OpenstatAdID String,  OpenstatSourceID String,  UTMSource String, UTMMedium String,
    UTMCampaign String,  UTMContent String,  UTMTerm String, FromTag String,  HasGCLID UInt8,  RefererHash UInt64,
    URLHash UInt64,  CLID UInt32,  YCLID UInt64,  ShareService String,  ShareURL String,  ShareTitle String,
    ParsedParams Nested(Key1 String,  Key2 String, Key3 String, Key4 String, Key5 String,  ValueDouble Float64),
    IslandID FixedString(16),  RequestNum UInt32,  RequestTry UInt8) ENGINE = MergeTree() PARTITION BY toYYYYMM(EventDate)
    ORDER BY (CounterID, EventDate, intHash32(UserID)) SAMPLE BY intHash32(UserID) SETTINGS index_granularity = 8192, storage_policy='$TEMP_POLICY'"
clickhouse-client --query "CREATE TABLE test.hits (WatchID UInt64,  JavaEnable UInt8,  Title String,  GoodEvent Int16,
    EventTime DateTime,  EventDate Date,  CounterID UInt32,  ClientIP UInt32,  ClientIP6 FixedString(16),  RegionID UInt32,
    UserID UInt64,  CounterClass Int8,  OS UInt8,  UserAgent UInt8,  URL String,  Referer String,  URLDomain String,
    RefererDomain String,  Refresh UInt8,  IsRobot UInt8,  RefererCategories Array(UInt16),  URLCategories Array(UInt16),
    URLRegions Array(UInt32),  RefererRegions Array(UInt32),  ResolutionWidth UInt16,  ResolutionHeight UInt16,  ResolutionDepth UInt8,
    FlashMajor UInt8, FlashMinor UInt8,  FlashMinor2 String,  NetMajor UInt8,  NetMinor UInt8, UserAgentMajor UInt16,
    UserAgentMinor FixedString(2),  CookieEnable UInt8, JavascriptEnable UInt8,  IsMobile UInt8,  MobilePhone UInt8,
    MobilePhoneModel String,  Params String,  IPNetworkID UInt32,  TraficSourceID Int8, SearchEngineID UInt16,
    SearchPhrase String,  AdvEngineID UInt8,  IsArtifical UInt8,  WindowClientWidth UInt16,  WindowClientHeight UInt16,
    ClientTimeZone Int16,  ClientEventTime DateTime,  SilverlightVersion1 UInt8, SilverlightVersion2 UInt8,  SilverlightVersion3 UInt32,
    SilverlightVersion4 UInt16,  PageCharset String,  CodeVersion UInt32,  IsLink UInt8,  IsDownload UInt8,  IsNotBounce UInt8,
    FUniqID UInt64,  HID UInt32,  IsOldCounter UInt8, IsEvent UInt8,  IsParameter UInt8,  DontCountHits UInt8,  WithHash UInt8,
    HitColor FixedString(1),  UTCEventTime DateTime,  Age UInt8,  Sex UInt8,  Income UInt8,  Interests UInt16,  Robotness UInt8,
    GeneralInterests Array(UInt16), RemoteIP UInt32,  RemoteIP6 FixedString(16),  WindowName Int32,  OpenerName Int32,
    HistoryLength Int16,  BrowserLanguage FixedString(2),  BrowserCountry FixedString(2),  SocialNetwork String,  SocialAction String,
    HTTPError UInt16, SendTiming Int32,  DNSTiming Int32,  ConnectTiming Int32,  ResponseStartTiming Int32,  ResponseEndTiming Int32,
    FetchTiming Int32,  RedirectTiming Int32, DOMInteractiveTiming Int32,  DOMContentLoadedTiming Int32,  DOMCompleteTiming Int32,
    LoadEventStartTiming Int32,  LoadEventEndTiming Int32, NSToDOMContentLoadedTiming Int32,  FirstPaintTiming Int32,
    RedirectCount Int8, SocialSourceNetworkID UInt8,  SocialSourcePage String,  ParamPrice Int64, ParamOrderID String,
    ParamCurrency FixedString(3),  ParamCurrencyID UInt16, GoalsReached Array(UInt32),  OpenstatServiceName String,
    OpenstatCampaignID String,  OpenstatAdID String,  OpenstatSourceID String,  UTMSource String, UTMMedium String,
    UTMCampaign String,  UTMContent String,  UTMTerm String, FromTag String,  HasGCLID UInt8,  RefererHash UInt64,
    URLHash UInt64,  CLID UInt32,  YCLID UInt64,  ShareService String,  ShareURL String,  ShareTitle String,
    ParsedParams Nested(Key1 String,  Key2 String, Key3 String, Key4 String, Key5 String,  ValueDouble Float64),
    IslandID FixedString(16),  RequestNum UInt32,  RequestTry UInt8) ENGINE = MergeTree() PARTITION BY toYYYYMM(EventDate)
    ORDER BY (CounterID, EventDate, intHash32(UserID)) SAMPLE BY intHash32(UserID) SETTINGS index_granularity = 8192, storage_policy='$TEMP_POLICY'"
clickhouse-client --query "CREATE TABLE test.visits (CounterID UInt32,  StartDate Date,  Sign Int8,  IsNew UInt8,
    VisitID UInt64,  UserID UInt64,  StartTime DateTime,  Duration UInt32,  UTCStartTime DateTime,  PageViews Int32,
    Hits Int32,  IsBounce UInt8,  Referer String,  StartURL String,  RefererDomain String,  StartURLDomain String,
    EndURL String,  LinkURL String,  IsDownload UInt8,  TraficSourceID Int8,  SearchEngineID UInt16,  SearchPhrase String,
    AdvEngineID UInt8,  PlaceID Int32,  RefererCategories Array(UInt16),  URLCategories Array(UInt16),  URLRegions Array(UInt32),
    RefererRegions Array(UInt32),  IsYandex UInt8,  GoalReachesDepth Int32,  GoalReachesURL Int32,  GoalReachesAny Int32,
    SocialSourceNetworkID UInt8,  SocialSourcePage String,  MobilePhoneModel String,  ClientEventTime DateTime,  RegionID UInt32,
    ClientIP UInt32,  ClientIP6 FixedString(16),  RemoteIP UInt32,  RemoteIP6 FixedString(16),  IPNetworkID UInt32,
    SilverlightVersion3 UInt32,  CodeVersion UInt32,  ResolutionWidth UInt16,  ResolutionHeight UInt16,  UserAgentMajor UInt16,
    UserAgentMinor UInt16,  WindowClientWidth UInt16,  WindowClientHeight UInt16,  SilverlightVersion2 UInt8,  SilverlightVersion4 UInt16,
    FlashVersion3 UInt16,  FlashVersion4 UInt16,  ClientTimeZone Int16,  OS UInt8,  UserAgent UInt8,  ResolutionDepth UInt8,
    FlashMajor UInt8,  FlashMinor UInt8,  NetMajor UInt8,  NetMinor UInt8,  MobilePhone UInt8,  SilverlightVersion1 UInt8,
    Age UInt8,  Sex UInt8,  Income UInt8,  JavaEnable UInt8,  CookieEnable UInt8,  JavascriptEnable UInt8,  IsMobile UInt8,
    BrowserLanguage UInt16,  BrowserCountry UInt16,  Interests UInt16,  Robotness UInt8,  GeneralInterests Array(UInt16),
    Params Array(String),  Goals Nested(ID UInt32, Serial UInt32, EventTime DateTime,  Price Int64,  OrderID String, CurrencyID UInt32),
    WatchIDs Array(UInt64),  ParamSumPrice Int64,  ParamCurrency FixedString(3),  ParamCurrencyID UInt16,  ClickLogID UInt64,
    ClickEventID Int32,  ClickGoodEvent Int32,  ClickEventTime DateTime,  ClickPriorityID Int32,  ClickPhraseID Int32,  ClickPageID Int32,
    ClickPlaceID Int32,  ClickTypeID Int32,  ClickResourceID Int32,  ClickCost UInt32,  ClickClientIP UInt32,  ClickDomainID UInt32,
    ClickURL String,  ClickAttempt UInt8,  ClickOrderID UInt32,  ClickBannerID UInt32,  ClickMarketCategoryID UInt32,  ClickMarketPP UInt32,
    ClickMarketCategoryName String,  ClickMarketPPName String,  ClickAWAPSCampaignName String,  ClickPageName String,  ClickTargetType UInt16,
    ClickTargetPhraseID UInt64,  ClickContextType UInt8,  ClickSelectType Int8,  ClickOptions String,  ClickGroupBannerID Int32,
    OpenstatServiceName String,  OpenstatCampaignID String,  OpenstatAdID String,  OpenstatSourceID String,  UTMSource String,
    UTMMedium String,  UTMCampaign String,  UTMContent String,  UTMTerm String,  FromTag String,  HasGCLID UInt8,  FirstVisit DateTime,
    PredLastVisit Date,  LastVisit Date,  TotalVisits UInt32,  TraficSource    Nested(ID Int8,  SearchEngineID UInt16, AdvEngineID UInt8,
    PlaceID UInt16, SocialSourceNetworkID UInt8, Domain String, SearchPhrase String, SocialSourcePage String),  Attendance FixedString(16),
    CLID UInt32,  YCLID UInt64,  NormalizedRefererHash UInt64,  SearchPhraseHash UInt64,  RefererDomainHash UInt64,  NormalizedStartURLHash UInt64,
    StartURLDomainHash UInt64,  NormalizedEndURLHash UInt64,  TopLevelDomain UInt64,  URLScheme UInt64,  OpenstatServiceNameHash UInt64,
    OpenstatCampaignIDHash UInt64,  OpenstatAdIDHash UInt64,  OpenstatSourceIDHash UInt64,  UTMSourceHash UInt64,  UTMMediumHash UInt64,
    UTMCampaignHash UInt64,  UTMContentHash UInt64,  UTMTermHash UInt64,  FromHash UInt64,  WebVisorEnabled UInt8,  WebVisorActivity UInt32,
    ParsedParams    Nested(Key1 String,  Key2 String,  Key3 String,  Key4 String, Key5 String, ValueDouble    Float64),
    Market Nested(Type UInt8, GoalID UInt32, OrderID String,  OrderPrice Int64,  PP UInt32,  DirectPlaceID UInt32,  DirectOrderID  UInt32,
    DirectBannerID UInt32,  GoodID String, GoodName String, GoodQuantity Int32,  GoodPrice Int64),  IslandID FixedString(16))
    ENGINE = CollapsingMergeTree(Sign) PARTITION BY toYYYYMM(StartDate) ORDER BY (CounterID, StartDate, intHash32(UserID), VisitID)
    SAMPLE BY intHash32(UserID) SETTINGS index_granularity = 8192, storage_policy='$TEMP_POLICY'"

# Might fail in sanitizer runs, not very important
set +e
clickhouse-client --max_execution_time 600 --max_memory_usage 30G --max_memory_usage_for_user 30G --query "INSERT INTO test.hits_s3 SELECT * FROM datasets.hits_v1 SETTINGS enable_filesystem_cache_on_write_operations=0, max_insert_threads=16"
clickhouse-client --max_execution_time 600 --max_memory_usage 30G --max_memory_usage_for_user 30G --query "INSERT INTO test.hits SELECT * FROM datasets.hits_v1 SETTINGS enable_filesystem_cache_on_write_operations=0, max_insert_threads=16"
clickhouse-client --max_execution_time 600 --max_memory_usage 30G --max_memory_usage_for_user 30G --query "INSERT INTO test.visits SELECT * FROM datasets.visits_v1 SETTINGS enable_filesystem_cache_on_write_operations=0, max_insert_threads=16"

clickhouse-client --query "DROP TABLE datasets.visits_v1 SYNC"
clickhouse-client --query "DROP TABLE datasets.hits_v1 SYNC"
# Drop `tpch` before the storage policy switch below. Its tables live on the `default` disk which becomes unavailable under
# `azure_cache`/`s3_cache`, preventing the server from starting. `tpcds` is not dropped because web disk survives policy changes.
clickhouse-client --query "DROP DATABASE IF EXISTS tpch SYNC"

clickhouse-client --query "SHOW TABLES FROM test"
set -e


stop_server

# Let's enable S3 storage by default
export RANDOMIZE_OBJECT_KEY_TYPE=1
export ZOOKEEPER_FAULT_INJECTION=1
# THREAD_POOL_FAULT_INJECTION is not exported here: if `cannot_allocate_thread_injection.xml`
# is installed before the server starts, the smoke-check `CREATE DATABASE ... ON CLUSTER`
# can hit `CANNOT_SCHEDULE_TASK` and break the HTTP response. `stress.py` installs the
# config and `SYSTEM RELOAD CONFIG`s the server after the smoke check passes.
export CLICKHOUSE_FAILPOINTS_INJECTION=1
configure
configure_limits

if [[ "$USE_S3_STORAGE_FOR_MERGE_TREE" == "1" || "$USE_AZURE_STORAGE_FOR_MERGE_TREE" == "1" ]]; then
    # But we still need default disk because some tables loaded only into it
    if [[ $USE_S3_STORAGE_FOR_MERGE_TREE == "1" ]]; then
        if [[ $USE_ENCRYPTED_STORAGE == "1" ]]; then
            file=/etc/clickhouse-server/config.d/s3_encrypted_storage_policy_for_merge_tree_by_default.xml
        else
            file=/etc/clickhouse-server/config.d/s3_storage_policy_by_default.xml
        fi
    elif [[ "$USE_AZURE_STORAGE_FOR_MERGE_TREE" == "1" ]]; then
        if [[ $USE_ENCRYPTED_STORAGE == "1" ]]; then
            file=/etc/clickhouse-server/config.d/azure_encrypted_storage_policy_by_default.xml
        else
            file=/etc/clickhouse-server/config.d/azure_storage_policy_by_default.xml
        fi
    else
        echo "ERROR: Failed to find azure storage policy by default file"
        exit 1
    fi

    sed -i.tmp "s|<main><disk>cached_azure</disk></main>|<main><disk>cached_azure</disk></main><default><disk>default</disk></default>|" "$file"

    sudo chown clickhouse "$file"
    sudo chgrp clickhouse "$file"
fi


# Disable experimental transactions support.
# TODO: Enable back after the issue with `assertHasValidVersionMetadata` will be fixed:
# https://play.clickhouse.com/play?user=play&run=1#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
rm -f /etc/clickhouse-server/config.d/transactions.xml

sed -i.tmp "s|<level>trace</level>|<level>test</level>|" /etc/clickhouse-server/config.d/logger_trace.xml

if [ "$cache_policy" = "SLRU" ]; then
    sed -i.tmp "s|<cache_policy>LRU</cache_policy>|<cache_policy>SLRU</cache_policy>|" /etc/clickhouse-server/config.d/storage_conf*.xml
fi

# Randomize async_load_databases
if [ $((RANDOM % 2)) -eq 0 ]; then
    sudo echo "<datastore><async_load_databases>false</async_load_databases></datastore>" \
        > /etc/clickhouse-server/config.d/enable_async_load_databases.xml
fi

# Randomize concurrent_threads_scheduler (default is max_min_fair)
if [ $((RANDOM % 2)) -eq 1 ]; then
    sudo echo "<datastore><concurrent_threads_scheduler>fair_round_robin</concurrent_threads_scheduler></datastore>" \
        > /etc/clickhouse-server/config.d/enable_max_min_fair_scheduler.xml
fi

# Cap the recursion-controlling parser/AST settings. The stress test ignores `no-*` build
# tags and runs the query fuzzer over every query, so deeply nested expressions (e.g. test
# 04412's formatQuery of a 20000-deep array literal) would otherwise build huge queries that
# hang the server under sanitizers and trip the hung check.
cp -av --dereference /repo/ci/jobs/scripts/fuzzer/limit-recursion-settings.xml /etc/clickhouse-server/users.d/

start_server || { echo "Failed to start server"; exit 1; }

cd /repo/tests/ || exit 1  # datastore-test can find queries dir from there
python3 /repo/ci/jobs/scripts/stress/stress.py --hung-check --drop-databases --output-folder /test_output --skip-func-tests "$SKIP_TESTS_OPTION" --global-time-limit "${STRESS_GLOBAL_TIME_LIMIT:-1200}" --encrypted-storage "$USE_ENCRYPTED_STORAGE" \
    && echo -e "Test script exit code$OK" >> /test_output/test_results.tsv \
    || echo -e "Test script failed$FAIL script exit code: $?" >> /test_output/test_results.tsv

stop_server
mv /var/log/clickhouse-server/clickhouse-server.log /var/log/clickhouse-server/clickhouse-server.stress.log

# NOTE Disable thread fuzzer before server start with data after stress test.
# In debug build it can take a lot of time.
unset "${!THREAD_@}"
# Disable cannot_allocate_thread_fault_injection_probability so the post-stress
# restart can load tables asynchronously. `-f` covers the case where the smoke
# check aborted before stress.py installed the symlink.
rm -f /etc/clickhouse-server/config.d/cannot_allocate_thread_injection.xml
rm -f /etc/clickhouse-server/config.d/fail_points_active.xml

# Use a larger timeout for the post-stress restart: under sanitizers with
# async_load_databases=false the server may need minutes to load all tables.
start_server 10 || { echo "Failed to start server"; exit 1; }

check_server_start

stop_server

[ -f /var/log/clickhouse-server/clickhouse-server.log ] || echo -e "Server log does not exist\tFAIL"
[ -f /var/log/clickhouse-server/stderr.log ] || echo -e "Stderr log does not exist\tFAIL"

mv /var/log/clickhouse-server/clickhouse-server.log /var/log/clickhouse-server/clickhouse-server.final.log

# Grep logs for sanitizer asserts, crashes and other critical errors
check_logs_for_critical_errors

tar -chf /test_output/coordination.tar /var/lib/clickhouse/coordination ||:

collect_query_and_trace_logs

mv /var/log/clickhouse-server/stderr.log /test_output/