Php fpm pool manager Static vs dynamic vs ondemand
; static –
a fixed number (pm.max_children) of child processes;
; dynamic –
the number of child processes are set dynamically based on the
; following directives. With this process management, there will be
; always at least 1 children.
; pm.max_children – the maximum number of children that can
; be alive at the same time.
; pm.start_servers – the number of children created on startup.
; pm.min_spare_servers – the minimum number of children in ‘idle’
; state (waiting to process). If the number
; of ‘idle’ processes is less than this
; number then some children will be created.
; pm.max_spare_servers – the maximum number of children in ‘idle’
; state (waiting to process). If the number
; of ‘idle’ processes is greater than this
; number then some children will be killed.
;
ondemand –
no children are created at startup. Children will be forked when
; new requests will connect. The following parameter are used:
; pm.max_children – the maximum number of children that
; can be alive at the same time.
; pm.process_idle_timeout – The number of seconds after which
; an idle process will be killed.
pm = Static, Dynamic, on demand
NOTE: apache mpm module configuration similar to this.
php fpm Static pool manager
pm.max_children = 5
php – fpm pool manager Dynamic
Default www.conf file
pm.max_children = 5
pm.start_servers = 2
pm.min_spare_servers = 1
pm.max_spare_servers = 3 (1+2)
pm.starts servers
min_spare_servers + (max_spare_servers – min_spare_servers) / 2
1+(3-1)/2=2
2/2=1 +1 =2.
//note less than min spare
ondemand;
pm.process_idle_timeout = 10s;
memory leaks safety
; Default Value: 0
;pm.max_requests = 500
Process has to dire or respawn after handling above request this way we avoid memory leaks.
respawning frequently also utilizes more memory so set it 10K.
ondemand 10s limit to kill the process upon idle.
; Set open file descriptor rlimit.
; Default Value: system defined value
;rlimit_files = 1024
; Set max core size rlimit.
; Possible Values: ‘unlimited’ or an integer greater or equal to 0
; Default Value: system defined value
;rlimit_core = 0
pm max children reached
fixing php-fpm pool high cpu usage
available memory in LEMP stack
4GB
1GB for MySQL, 512MB for linux, 1GB for nginx
2.5GB (1.5GB,
php max memory limit (128MB or 256MB)
max execution time 60s or 300s 1 minute or 5 minutes
slow query log threshold 1s.
starart severs +idle servers
all are max child processes. handles only one connection per second.
30 connection per second.
30*128MB=3048M (but all requests not consumes 128M)
4MB or 8MB average.
monitor slow query logs.
static: pool manager
when the limit is 30.
30*128M = 3GB Max dedicated needed.
minimum memory ==?
each process occupies memory to handle requests even though no request there
dynamic for mid range request
On demand for (free memory)
dynamic is better holds 3 max spare & total 5 spare
we have increase upto 10. If you get message like
max children reached.
enable php fpm slow query log
slowlog = /var/log/php-slow.log
request_slowlog_timeout = 10s
or
slowlog = /var/log/$pool.log.slow
equest_slowlog_timeout = 10s
uncomment the lines in /etc/php/7.3/fpm/pool.d/www.conf
Deciding the Slow query threshold if you updates wordpress plugins it takes times,
if you upload 8MB files its takes time.
Best php query time for wordpress is ….? (wordpress load in 1-3 secs) (php execution time time , request terminate timeout ).
Monitor the script and adjust the value.
restart or reload php fpm
root@instance-1:~#
systemctl restart php7.3-fpm
service reload php7.3-fpm
also read enabling mysql slow query log
enable php from status page to better understand
Step 1: uncomment pm /status
step 2: add nginx server block
for understanding php-fpm status page here.
Monitoring Requests in with nginx