I've been getting this error when logged in users try to create new posts
, save new posts
upstream timed out (110: Connection timed out) while reading response header from upstream, client: 127.0.0.1, server: www.website.com, request: "POST /wp-admin/admin-ajax.php HTTP/1.1", upstream: "fastcgi://127.0.0.1:9000", host: "www.website.com", referrer: "http://www.website.com/wp-admin/post-new.php"
After racking my brains and trying almost everything I can find online, I see that mysql is taking too much cpu and it's in sleep state:
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1165 mysql 20 0 6710000 392500 7704 S 538.4 1.2 429:15.23 mysqld
8538 nginx 20 0 1011120 767704 28384 R 96.5 2.3 0:38.88 php5-fpm
8631 nginx 20 0 1008752 765972 28516 S 64.8 2.3 0:41.22 php5-fpm
8933 nginx 20 0 342324 105612 27164 S 14.9 0.3 1:12.39 php5-fpm
8282 nginx 20 0 284956 47704 28680 S 11.9 0.1 0:22.16 php5-fpm
8696 nginx 20 0 285428 48580 27180 S 11.9 0.1 1:06.20 php5-fpm
7893 nginx 20 0 276920 39524 28536 S 9.9 0.1 0:29.77 php5-fpm
8839 nginx 20 0 284152 47772 27736 S 9.9 0.1 0:07.72 php5-fpm
1239 nobody 20 0 1562720 388788 83312 S 8.9 1.2 9:01.75 varnishd
9176 nginx 20 0 282612 43620 24884 S 8.9 0.1 0:00.95 php5-fpm
7781 nginx 20 0 287504 50200 28740 S 8.3 0.2 0:39.74 php5-fpm
8905 nginx 20 0 283128 47108 27852 S 8.3 0.1 0:05.36 php5-fpm
7913 nginx 20 0 284684 47424 28672 S 7.6 0.1 0:29.60 php5-fpm
8728 nginx 20 0 587692 350692 27232 S 7.6 1.1 1:06.50 php5-fpm
8750 nginx 20 0 282872 46640 27644 S 7.6 0.1 0:08.28 php5-fpm
8277 nginx 20 0 313220 75636 28504 S 6.9 0.2 1:19.49 php5-fpm
8675 nginx 20 0 284944 47280 28268 S 6.9 0.1 0:09.52 php5-fpm
How to solve this issue?
I'm not able to save anything or create new posts, the dashboard is horribly slow. Please guide me.
UPDATE: I decided to check the wait_timeout and this is what I got:
SHOW SESSION VARIABLES LIKE "%wait%";
+---------------------------------------------------+----------+
| Variable_name | Value |
+---------------------------------------------------+----------+
| innodb_lock_wait_timeout | 50 |
| innodb_spin_wait_delay | 6 |
| lock_wait_timeout | 31536000 |
| performance_schema_events_waits_history_long_size | 10000 |
| performance_schema_events_waits_history_size | 10 |
| wait_timeout | 28800 |
+---------------------------------------------------+----------+
Is it just me or the wait_timeout
of 28800 is insanely high?
Can this be the problem?