0

Store run on Linode 8 GB RAM 4 cores, Magento 2, Nginx, PHP-FPM 7.2. I ran a test with 1000 clients for a minute and I see high error rate. Not much in magento or nginx error logs. Guidance please?

pm = dynamic
pm.max_children = 125
pm.max_requests = 500
pm.start_servers = 37
pm.min_spare_servers = 25
pm.max_spare_servers = 75

innodb_buffer_pool_size = 4G
innodb_log_file_size = 256M
innodb_flush_log_at_trx_commit = 1 # may change to 2 or 0
innodb_flush_method = O_DIRECT
innodb_buffer_pool_instances = 4
innodb_thread_concurrency = 8
innodb_io_capacity = 1000
innodb_read_io_threads = 8
innodb_write_io_threads = 8
innodb_lru_scan_depth=256

enter image description here

4
  • Magento is a resource hungry beast. What tuning have you done? Jan 22, 2019 at 14:17
  • I have copied and pasted above the tuning that I have done for PHP-FPM and percona db. I also have redis configured.
    – Zoya
    Jan 22, 2019 at 14:37
  • You're meant to be collecting metrics on how your server is performing, especially during a load test. What happened on your server when you ran the test? Jan 22, 2019 at 14:49
  • In Linode 4 cores means 400% CPU allocated. I see that CPU doesn't even crosses 100% for the first core. Rest of the 3 cores aren't being used I believe. With regards to memory I still see 2-3 G available while running the test
    – Zoya
    Jan 22, 2019 at 15:07

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.