9

Been sitting on this for 12 hours straight (it's noon now, so I will read/respond when I woke up again).
I have made the big fault of recommending a database upgrade of our productive environment to increase performance.
We can NOT go back, it's nearly 6 terrabytes of storage and downgrades are not possible with mysql 8.0.
Also using a pre upgrade snapshot backup is no solution (days of work).

I would like to solve the horrible performance that resulted from switching server to latest version

A few details:
Environment: Linux Stretch on AWS i3.8xlarge (32 cpus, 240gb ram)
Server: Mysql 8.0 in mysql/ Docker container using bind mounts and host networking
Storage: Amazon AWS EBS IO1 storage (6 TB) with 20,000 IOPS reserved.
The storage delivers 500mb/sec-600mb/sec in fio tests.
CPU: Usually the 32 cores are 50-60% used but since mysql 8.0 it's idle (10-15% used)
RAM: 200GB are dedicated to mysql, before 8.0 it took not long and it was used. Now it takes many hours until mysql is able to fill the buffer.

The core problem: The speed dropped about 20 times in comparison to 5.7
Innodb/mysql is not using the disk efficiently.
Using the original configuration it was reading with 15 mb/sec (simple counts took minutes to run because the IBD file was not read properly)
I have since disabled performance schema, which helped to at least operate the server again at 10% load.
I disabled bin logging, maybe that helped a bit more, not sure.
I tried increasing read/write threads to 64 (that completely stalled the server)

I invested good 8 hours in trying to tune the mysql configuration, I managed to "push" mysql from using only 5mb/sec to now up to 50mb/sec.

To make sure:
Of course I did test disk IO from inside docker, it's exactly similar to outside.
The main disk is mostly idle, the other disks completely idle.
The system is 90% idle when it should be occupied.

Innodb status:
https://pastebin.com/XDgyNbk0

Mysql configuration:

[mysqld]
user=mysql
pid-file        = /var/run/mysqld/mysqld.pid
socket          = /var/run/mysqld/mysqld.sock
datadir         = /var/lib/mysql

tmpdir          = /mysql_tmp  

# compatibility
default_authentication_plugin=mysql_native_password
character_set_server=latin1
collation_server=latin1_swedish_ci



log-error       = /var/log/mysql_error.log
bind-address    = 0.0.0.0
sql_mode= "NO_ENGINE_SUBSTITUTION"

interactive_timeout     = 3600
wait_timeout            = 900

max_allowed_packet      = 64M
thread_stack            = 256K
thread_cache_size       = 192

max_connections         = 1600
max_user_connections    = 1500

#query_cache_limit       = 3M
#query_cache_size        = 200M
#query_cache_type       = 1
table_open_cache        = 2500


key_buffer_size          = 64M   # index in memory for myisam
innodb_buffer_pool_size = 190G
innodb_log_file_size = 256M
tmp_table_size = 250M
max_heap_table_size = 250M
join_buffer_size = 2M
#pagecleaners  - those were uncommented on 5.7
#innodb_buffer_pool_instances=8
#innodb_page_cleaners=2
innodb_io_capacity=5000
innodb_io_capacity_max=20000

# tried 64, that totally stalled the database
innodb_read_io_threads = 8
innodb_write_io_threads = 8

#in pre 5.7 times I had consistent 300mb/sec writes, now it's useless
innodb_lru_scan_depth=256

skip-name-resolve

secure_file_priv=""
#innodb_checksum_algorithm = crc32
#binlog_checksum = CRC32

# this one at least made it possible so I can go to bed, with performance_schema the database was unuseable    
performance_schema=OFF
skip-log-bin 

Sysctl fs:

fs.aio-max-nr = 1048576
fs.aio-nr = 139264
fs.binfmt_misc.status = enabled
fs.dentry-state = 355223        335269  45      0       0       0
fs.dir-notify-enable = 1
fs.epoll.max_user_watches = 51660308
fs.file-max = 25224638
fs.file-nr = 19136      0       25224638
fs.inode-nr = 70145     5686
fs.inode-state = 70145  5686    0       0       0       0       0
fs.inotify.max_queued_events = 16384
fs.inotify.max_user_instances = 128
fs.inotify.max_user_watches = 8192
fs.lease-break-time = 45
fs.leases-enable = 1
fs.mqueue.msg_default = 10
fs.mqueue.msg_max = 10
fs.mqueue.msgsize_default = 8192
fs.mqueue.msgsize_max = 8192
fs.mqueue.queues_max = 256
fs.nr_open = 1048576
fs.overflowgid = 65534
fs.overflowuid = 65534
fs.pipe-max-size = 1048576
fs.protected_hardlinks = 1
fs.protected_symlinks = 1
fs.quota.allocated_dquots = 0
fs.quota.cache_hits = 0
fs.quota.drops = 0
fs.quota.free_dquots = 0
fs.quota.lookups = 0
fs.quota.reads = 0
fs.quota.syncs = 62
fs.quota.warnings = 1
fs.quota.writes = 0
fs.suid_dumpable = 0

iostat snapshot:

avg-cpu:  %user   %nice %system %iowait  %steal   %idle
          13.86    0.71   10.84    3.23    0.11   71.26

Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
xvdh           1795.31     13923.89     26159.95  256609017  482112504

As you can see it's just reading at 14mb/sec, writing at 26mb/sec ..
With mysql 5.7 it was doing up to 200mb reading and writing.

The disk is basically idle. It can deliver 10 times the performance but for an unknown reason innodb/mysql are not doing it anymore.

Update: Variables and global status: https://pastebin.com/pzhXV7hq
https://pastebin.com/jBTYLbY6

Another change I was forced to make:
I have a hundred users connecting per second through apache/php, every connection usually triggered a "SELECT count(*) FROM information_schema.processlist" The new mysql wasn't able to do that anymore, it queued 50+ of those selects so I instead made an asynchronous task that inserts the processlist into an innodb table every 5 seconds.
It's just another sign how less performant the new mysql is reacting, it's even choked by processlists.

ulimit

   core file size          (blocks, -c) 0
    data seg size           (kbytes, -d) unlimited
    scheduling priority             (-e) 0
    file size               (blocks, -f) unlimited
    pending signals                 (-i) 985342
    max locked memory       (kbytes, -l) 64
    max memory size         (kbytes, -m) unlimited
    open files                      (-n) 65536
    pipe size            (512 bytes, -p) 8
    POSIX message queues     (bytes, -q) 819200
    real-time priority              (-r) 0
    stack size              (kbytes, -s) 8192
    cpu time               (seconds, -t) unlimited
    max user processes              (-u) 985342
    virtual memory          (kbytes, -v) unlimited
    file locks                      (-x) unlimited

Big update
Spent half a day reading into internas, taking your suggestions in consideration and what I was told on IRC.
I did the opposite of the professional: I made 10 changes at once, I can't afford so many restarts in a gradual change process:
1) Giving the system lots of parallel write opportunity without choking it

innodb_read_io_threads = 16  
innodb_write_io_threads = 16  
innodb_thread_concurrency=64  # cpus*2  

2) Speeding up the background sync:

 innodb_lru_scan_depth=100  

3) Disabling highest reliability settings that have a hard performance impact

performance_schema=OFF
skip-log-bin 
sync_binlog=0 
innodb_flush_log_at_trx_commit=0  # not crash safe, 0 is crash safe  

4) More multithreading in backend memory

innodb_buffer_pool_instances=12  

5) Increasing logfiles significantly, increasing logfile buffer moderatly

innodb_log_file_size = 3G # 
innodb_log_buffer_size = 64M

What happened: About 10 times increase in read performance,1.5x times write performance, I am not where I want to be but it's 15 times faster than before!
IOPS usage doubled from ~5-6k to 9k-12k, so I am at 60% IO usage
CPU usage increased from 7% to 50%

My goal would be 80% IO and CPU usage by the database, I think other variables are bottlenecking.

In real time use: I have a huge insert running before and after change (next to the usual load).
Before change the speed was about 3000 rows per second, after the above changes it's 8000 rows per second.

Thought I share this as the change in performance is extreme and I've only reached 50% of what should be possible.

Update
I think the problem can be considered half solved, I've made another update after the previous successful one and the performance is acceptable now.

The last changes involved write/read threads. I've put them to 32 each.
write buffer increased to 128M (for my heavy workload higher might be better)
logfiles increased to 8GB
buffer_pool_instances increased to 64 (max) for better memory fragmentation
page_cleaners increased to 64 (max) to have one for each buffer instance.

write performance increased by another ~20%, read performance increased by another ~30%.

It's been a 24 hour ride to get mysql perform acceptable, definitely not a simple upgrade.

Latest status:

Current configuration: https://pastebin.com/9vsbEQxt
show engine status innodb: https://pastebin.com/kCjnmtze
show global variables: https://pastebin.com/aMdQxWcA
global status: https://pastebin.com/VbG1yzHX
25
  • 3
    Can you provide some examples of how MySQL is slower? Is it writes performing slower, reads being impacted the most or do you see both taking longer? What MySQL testing did you do prior to upgrades? How did you perform the upgrade? Have you performed a pt-query-digest on slow log or against the running process list since the upgrade and how does it compare to previous profiles?
    – eroomydna
    Jan 22, 2019 at 10:23
  • 1
    Kondybas: roughly 150mb/sec single threaded, but it can go to 600mb/sec when doing it a few times in parallel. @eroomydna: It's stalling, if I would use the similar amount of processes accessing the database it just doesn't keep up and stalls with them all queued up. DISK IO stays low at any time, it doens't read the files into memory. counting 1 million rows takes 13 seconds when it took 4 seconds before (no cache). inserting 20,000 rows into a larger table takes 90 seconds, it took 12 seconds before. Performance is a fraction of what it was
    – John
    Jan 22, 2019 at 18:18
  • 1
    @RickJames I have added global status and variables for V8, I can't deliver V57 anymore. However the configuration file was almost identical, the changed parts are read and write threads and the commented query cache stuff. It's important to see those values as "idle" state. Under usual performance the server was writing average 250-300 MB/second to the disk. Since V8 it's average 50mb/sec and that was only possible when I disabled performance schema and set trx_commit to 2. Usually I have 30 threads working on tables, now I have 5 threads.
    – John
    Jan 22, 2019 at 18:35
  • 1
    @mustaccio Previous one was not in a container, I am running Jessie Debian which doesn't support Mysql 8.0 without a libc upgrade that I won't risk. But Containers are a good option for mysql in any case. It is the official "mysql/mysql-server:8.0" docker image. It runs on the same EC2 instance (it's idle now, it was busy with 5.7) and same memory allocation. The container uses a BIND mount to the ebs drive, so there is no IO overhead, it's as fast as outside. Also no restrictions on cpu or anything.
    – John
    Jan 22, 2019 at 19:41
  • 1
    @RickJames Basically my core question is simple: What can I do to make mysql utilize the disks as much as possible. I have 20k IOPS and 600mb/sec but mysql only uses 8-10% of that. I have 32 cpu cores, only 7% are used. It was good with 5.7. but that's spilt milk, I just want to make mysql USE the resources it has available. It sits on the idly
    – John
    Jan 22, 2019 at 19:50

3 Answers 3

4

Review of GLOBAL STATUS and VARIABLES

Observations:

  • Version: 8.0.13
  • 240 GB of RAM
  • Uptime = 10:47:09; some GLOBAL STATUS values may not be meaningful yet.
  • You are not running on Windows.
  • Running 64-bit version
  • You appear to be running entirely (or mostly) InnoDB.

The More Important Issues:

Even though table_open_cache is high, it may be good to raise it further. Raising table_open_cache_instances to 32 may help, too.

innodb_buffer_pool_instances = 16; ditto for innodb_page_cleaners.

Increase innodb_log_file_size to 8G; see the 8.0 manual on details on how to achieve this (things have changed).

Why do you have innodb_flush_method = fsync? O_DIRECT is recommended in many situations.

Shrink long_quer_time; check the slowlog; there are some naughty queries we need to investigate.

CREATE TABLE every 5 seconds? What is going on? It is unusual to do more than a few such per day.

REPLACE is mostly supplanted by IODKU; what is your use case?

You are using various SHOW commands a lot. Their implementation has changed drastically in 8.0 and I would expect them to be faster in 8.0, but this could be one source of a slowdown.

10 TRUNCATE TABLE per minute? What's up? If you are replacing the contents of a table, it is usually better to create a new table, populate it, then use RENAME TABLE to atomically swap it into place.

Details and other observations:

( Innodb_buffer_pool_pages_flushed ) = 96,254,291 / 38829 = 2478 /sec -- Writes (flushes)

( Opened_tables ) = 104,524 / 38829 = 2.7 /sec -- Frequency of opening Tables -- increase table_open_cache

( table_open_cache ) = 19,000 -- Number of table descriptors to cache -- Several hundred is usually good.

( Table_open_cache_misses ) = 104,520 / 38829 = 2.7 /sec -- May need to increase table_open_cache

( innodb_buffer_pool_size / innodb_buffer_pool_instances ) = 181248M / 8 = 22656MB -- Size of each buffer_pool instance. -- An instance should be at least 1GB. In very large RAM, have 16 instances.

( innodb_page_cleaners / innodb_buffer_pool_instances ) = 4 / 8 = 0.5 -- page_cleaners -- Recommend setting innodb_page_cleaners to innodb_buffer_pool_instances

( (Innodb_buffer_pool_reads + Innodb_buffer_pool_pages_flushed) ) = ((23804167 + 96254291) ) / 38829 = 3091 /sec -- InnoDB I/O

( Innodb_os_log_written ) = 249,997,739,520 / 38829 = 6438428 /sec -- This is an indicator of how busy InnoDB is.

( Innodb_log_writes ) = 411,565,494 / 38829 = 10599 /sec

( Innodb_os_log_written / (Uptime / 3600) / innodb_log_files_in_group / innodb_log_file_size ) = 249,997,739,520 / (38829 / 3600) / 2 / 256M = 43.2 -- Ratio -- (see minutes)

( Uptime / 60 * innodb_log_file_size / Innodb_os_log_written ) = 38,829 / 60 * 256M / 249997739520 = 0.695 -- Minutes between InnoDB log rotations Beginning with 5.6.8, this can be changed dynamically; be sure to also change my.cnf. -- (The recommendation of 60 minutes between rotations is somewhat arbitrary.) Adjust innodb_log_file_size. (Cannot change in AWS.)

( innodb_flush_method ) = innodb_flush_method = fsync -- How InnoDB should ask the OS to write blocks. Suggest O_DIRECT or O_ALL_DIRECT (Percona) to avoid double buffering. (At least for Unix.) See chrischandler for caveat about O_ALL_DIRECT

( Innodb_row_lock_waits ) = 41,325 / 38829 = 1.1 /sec -- How often there is a delay in getting a row lock. -- May be caused by complex queries that could be optimized.

( Innodb_dblwr_writes ) = 1,177,834 / 38829 = 30 /sec -- "Doublewrite buffer" writes to disk. "Doublewrites" are a reliability feature. Some newer versions / configurations don't need them. -- (Symptom of other issues)

( Innodb_row_lock_current_waits ) = 54 -- The number of row locks currently being waited for by operations on InnoDB tables. Zero is pretty normal. -- Something big is going on?

( innodb_print_all_deadlocks ) = innodb_print_all_deadlocks = OFF -- Whether to log all Deadlocks. -- If you are plagued with Deadlocks, turn this on. Caution: If you have lots of deadlocks, this may write a lot to disk.

( Queries ) = 146,600,019 / 38829 = 3775 /sec -- Queries (including inside SP)

( Created_tmp_tables ) = 43,466,450 / 38829 = 1119 /sec -- Frequency of creating "temp" tables as part of complex SELECTs.

( Created_tmp_disk_tables ) = 69,721 / 38829 = 1.8 /sec -- Frequency of creating disk "temp" tables as part of complex SELECTs -- increase tmp_table_size and max_heap_table_size. Check the rules for temp tables on when MEMORY is used instead of MyISAM. Perhaps minor schema or query changes can avoid MyISAM. Better indexes and reformulation of queries are more likely to help.

( Select_full_join ) = 341,473 / 38829 = 8.8 /sec -- joins without index -- Add suitable index(es) to tables used in JOINs.

( Select_scan ) = 20,070,358 / 38829 = 516 /sec -- full table scans -- Add indexes / optimize queries (unless they are tiny tables)

( Select_scan / Com_select ) = 20,070,358 / 46219009 = 43.4% -- % of selects doing full table scan. (May be fooled by Stored Routines.) -- Add indexes / optimize queries

( Com_insert + Com_delete + Com_delete_multi + Com_replace + Com_update + Com_update_multi ) = (1395278 + 512340 + 0 + 1843851 + 14825066 + 9554) / 38829 = 478 /sec -- writes/sec -- 50 writes/sec + log flushes will probably max out I/O write capacity of normal drives

( Com_replace ) = 1,843,851 / 38829 = 47 /sec -- Consider changing to INSERT ... ON DUPLICATE KEY UPDATE.

( expire_logs_days ) = 0 -- How soon to automatically purge binlog (after this many days) -- Too large (or zero) = consumes disk space; too small = need to respond quickly to network/machine crash. (Not relevant if log_bin = OFF)

( slave_pending_jobs_size_max / max_allowed_packet ) = 128M / 64M = 2 -- For parallel slave threads -- slave_pending_jobs_size_max must not be less than max_allowed_packet

( long_query_time ) = 10 -- Cutoff (Seconds) for defining a "slow" query. -- Suggest 2

( back_log / max_connections ) = 1,600 / 1600 = 100.0%

( Connections ) = 1,054,868 / 38829 = 27 /sec -- Connections -- Increase wait_timeout; use pooling?

( thread_cache_size ) = 192 -- How many extra processes to keep around (Not relevant when using thread pooling) (Autosized as of 5.6.8; based on max_connections)

Abnormally large:

Bytes_received = 1583925 /sec
Com_begin = 11 /sec
Com_create_db = 0.093 /HR
Com_create_function = 4.1 /HR
Com_create_procedure = 0.37 /HR
Com_create_table = 0.21 /sec
Com_delete = 13 /sec
Com_drop_procedure = 0.37 /HR
Com_insert_select = 29 /sec
Com_insert_select + Com_replace_select = 75 /sec
Com_kill = 0.74 /HR
Com_replace_select = 47 /sec
Com_show_create_db = 2.2 /HR
Com_show_create_trigger = 0.28 /HR
Com_show_events = 0.74 /HR
Com_show_storage_engines = 0.93 /HR
Com_show_warnings = 16 /HR
Com_stmt_close = 919 /sec
Com_stmt_execute = 919 /sec
Com_stmt_prepare = 919 /sec
Com_truncate = 0.16 /sec
Com_update = 381 /sec
Com_update_multi = 0.25 /sec
Handler_commit = 1638 /sec
Handler_delete = 60 /sec
Handler_external_lock = 5083 /sec
Handler_read_key = 58652 /sec
Handler_read_next = 607861 /sec
Handler_update = 40571 /sec
Innodb_buffer_pool_bytes_dirty = 1,309.3MB
Innodb_buffer_pool_pages_data = 1.18e+7
Innodb_buffer_pool_pages_dirty = 83,894
Innodb_buffer_pool_pages_flushed / max(Questions, Queries) = 0.657
Innodb_buffer_pool_pages_misc = 664,249
Innodb_buffer_pool_pages_total = 1.25e+7
Innodb_data_fsyncs = 232 /sec
Innodb_data_pending_fsyncs = 0.46 /HR
Innodb_data_read = 12400226 /sec
Innodb_data_reads = 770 /sec
Innodb_data_writes = 13131 /sec
Innodb_data_written = 88604121 /sec
Innodb_dblwr_pages_written = 2476 /sec
Innodb_dblwr_pages_written / Innodb_dblwr_writes = 81.6
Innodb_log_write_requests = 10906 /sec
Innodb_os_log_fsyncs = 87 /sec
Innodb_os_log_written / (Uptime / 3600) / innodb_log_files_in_group = 11,052.3MB
Innodb_pages_created = 57 /sec
Innodb_pages_written = 2478 /sec
Innodb_rows_deleted = 60 /sec
Innodb_rows_deleted + Innodb_rows_inserted = 658 /sec
Innodb_rows_inserted = 598 /sec
Innodb_rows_updated = 520 /sec
Max_execution_time_set = 0.0MB
Max_used_connections = 768
Select_range = 1250 /sec
Select_range / Com_select = 105.0%
Sort_range = 274 /sec
Sort_scan = 509 /sec
Table_open_cache_hits = 3826 /sec
Threads_cached = 142
Threads_connected = 366
back_log = 1,600
innodb_io_capacity_max = 20,000
innodb_max_dirty_pages_pct_lwm = 1000.0%
innodb_undo_tablespaces = 2
max_error_count = 1,024
max_length_for_sort_data = 4,096
max_user_connections = 1,500
optimizer_trace_max_mem_size = 1.05e+6
slave_pending_jobs_size_max = 128MB

Abnormal strings:

bind_address = 0.0.0.0
event_scheduler = ON
explicit_defaults_for_timestamp = ON
have_query_cache = NO
have_ssl = YES
have_symlink = DISABLED
innodb_buffer_pool_dump_at_shutdown = ON
innodb_buffer_pool_load_at_startup = ON
innodb_fast_shutdown = 1
innodb_undo_directory = ./
innodb_undo_log_truncate = ON
master_info_repository = TABLE
optimizer_trace = enabled=off,one_line=off
optimizer_trace_features = greedy_search=on, range_optimizer=on, dynamic_range=on, repeated_subselect=on
relay_log_info_repository = TABLE
slave_rows_search_algorithms = INDEX_SCAN,HASH_SCAN
ssl_ca = ca.pem
ssl_cert = server-cert.pem
ssl_key = server-key.pem
transaction_write_set_extraction = XXHASH64
17
  • First of all thanks for taking so much time. I'm going through all you wrote once again. The problem is a LOT has changed since my question, I invested day and night into optimizing (some queries that relied on query_cache as well as tons of settings) I will update a new pastebin here in a sec, many of your suggestions are partly in there already. Did you analyze that all by hand or do you have a tool for that ?
    – John
    Jan 24, 2019 at 4:42
  • Current configuration: pastebin.com/9vsbEQxt show engine status innodb: pastebin.com/kCjnmtze show global variables: pastebin.com/aMdQxWcA That's the current situation. One main difference is that I have not 16 but 64 buffer instances. Not sure if that's bad ? (still much more than 1 GB per instance) The performance is roughtly 10-15 times better than in beginning but still not great. I've started moving a lot of logic into MEMCACHED to avoid mysql.
    – John
    Jan 24, 2019 at 4:54
  • @John - Both. First, a took does all the grunt work of, for example, dividing by Uptime to get "per second" values. And it is years of gathering tips to say what is "too big" or "too small". Then it is manual -- to toss cruft, remove contradictions, etc., plus further interpret the advice for your situation. I learn a thing or two in the process.
    – Rick James
    Jan 24, 2019 at 4:54
  • @John - As far as I know, buffer_pool_instances helps only a small percentage. One source recommends no more than 16, but with no explanation of why. I consider that setting to be relatively minor, and I suspect 64 to be OK, but well into "diminishing returns".
    – Rick James
    Jan 24, 2019 at 4:56
  • 2
    There may be mileage in limiting the number of threads inside the InnoDB kernel using innodb-thread-concurrency. This is set to 0 (unlimited) currently, according to your global vars list, and by lowering the amount of threads working at once may permit more threads to complete their business before context switching. Left unbounded can cause cpu saturation.
    – eroomydna
    Jan 24, 2019 at 6:40
1
UPDATE  task_s_b SET result_delivered=1, result_data='DISABLED',
        result_gathered=1
    WHERE  result_data is NULL
      AND  (assigned_counter >= 4
              OR  c_counter > 2
           ) 

Depending on the distribution of the values in that table, this query may be causing troubles.

Consider rewriting as two queries:

UPDATE  task_s_b SET result_delivered=1, result_data='DISABLED',
        result_gathered=1
    WHERE  result_data is NULL
      AND  assigned_counter >= 4;
UPDATE  task_s_b SET result_delivered=1, result_data='DISABLED',
        result_gathered=1
    WHERE  result_data is NULL
      AND  c_counter > 2;

And adding these two composite indexes with the columns in the specified order:

INDEX(result_data, assigned_counter)
INDEX(result_data, c_counter)

To discuss further, please provide SHOW CREATE TABLE and SHOW TABLE STATUS for that table.

This may cut back on deadlocks and speed up the server.

4
  • Looking at UPDATE task_sc_bi SET result_delivered=1, result_data='DISABLED', result_gathered=1 WHERE result_data is NULL AND (assigned_counter >= 2) -- Have you added the two indexes I recommended?
    – Rick James
    Jan 24, 2019 at 5:06
  • I can't add an index there at the moment. It would take 3-4 days to complete the operation :( I am currently adding indexes to another table, runs since 12 hours and I hope a lot that it will finish in the next 24 hours as I need the table ready again :(
    – John
    Jan 24, 2019 at 5:45
  • @John - I think the 8.0 version you are running has a faster ADD INDEX; check the docs.
    – Rick James
    Jan 24, 2019 at 15:25
  • @john In 8.0 refman at, dev.mysql.com/doc/refman/8.0/en/create-index.html - you will find CREATE INDEX syntax with ALGORITHM=INPLACE available to avoid the days long copy. Feb 2, 2019 at 21:30
0

Suggestions to consider with no stop/start of services for immediate relief based on available data at this time Jan 22, 19 at 15:00 CT USA

SET GLOBAL innodb_io_capacity=10000        from your 5000 limit at this time
SET GLOBAL innodb_lru_scan_depth=100        from your 256 to reduce CPU cycles used for this function by 60% EVERY SECOND

After posting of SHOW GLOBAL VARIABLES; analysis completed there will be more suggestions.

11
  • 2
    Hi, I've tried values from 100 to 20000, it didn't have a real impact on the performance observed. I'll try that combination. At the end of my question you'll find a pastebin with variables. The server is mostly idle, 7% CPU usage, 10% IO usage
    – John
    Jan 22, 2019 at 21:11
  • @John do you have flexibility to us Skype TALK with me? View profile, Network profile for contact info including my Skype ID for invite, please. I have picked up GV and am looking at analysis. Jan 22, 2019 at 21:51
  • 1
    I'm adding it. I was using fsync on 5.7 but switched to direct during my performance tests yesterday. So it's direct. I am adding ulimit too. I fell asleep, we can skype tomorrow if you have time. I only slept a few hours today after staying up till noon due to the issues. the database is "ok" in writing currently but really slow when accessing data outside innodb memory buffer. It is just not loading the IBD files fast, as if it was a an old hdd with rusty mechanics. I can load the whole IBD file in seconds with DD but mysql needs a day
    – John
    Jan 22, 2019 at 23:01
  • 1
    One more thing: 6mb/sec is low! I had 250mb/sec average write before the upgrade, 24 hours a day. When watching the logfiles (not mysql, those are empty) I can see that the processes which usually work very fast are getting stuck for seconds regularly. in processlist I can see that normal (highly optimized) queries are stuck for 6+ seconds. Then suddenly mysql works again and everything is fast for some seconds, then all is stalling again. It's all an IO issue I think, it's not accessing the disks properly.
    – John
    Jan 22, 2019 at 23:09
  • 2
    @John Thanks for your positive comment, even at 22 months after the event. You are appreciated. Nov 21, 2020 at 21:10

Not the answer you're looking for? Browse other questions tagged or ask your own question.