Fastdfs的tracker服务器出现一定概率的connection timeout

Question

@余庆 你好，想跟你请教个问题：

Fastdfs的tracker服务器在5000并发下，出现明显的响应时间周期性变长，然后出现大量的java.net.ConnectException: Connection timed out。此时服务器的内存和cpu均未跑满（配置：16核CPU，32G内存）

并且我们在较低并发（1000并发）的持久测试中也发现，有一定概率的java.net.ConnectException: Connection timed out。

tracker.conf如下：

disabled=false
bind_addr=
port=22122
connect_timeout=30
network_timeout=30
base_path=/home/yuqing/fastdfs
max_connections=10000
work_threads=32
store_lookup=0
store_group=group2
store_server=0
store_path=0
download_server=0
reserved_storage_space = 10%
log_level=debug
run_by_group=
run_by_user=
allow_hosts=*
sync_log_buff_interval = 10
check_active_interval = 120
thread_stack_size = 256KB
storage_ip_changed_auto_adjust = true
storage_sync_file_max_delay = 86400
storage_sync_file_max_time = 300
use_trunk_file = false 
slot_min_size = 256
slot_max_size = 16MB
trunk_file_size = 64MB
trunk_create_file_advance = false
trunk_create_file_time_base = 02:00
trunk_create_file_interval = 86400
trunk_create_file_space_threshold = 20G
trunk_init_check_occupying = false
trunk_init_reload_from_binlog = false
use_storage_id = false
storage_ids_filename = storage_ids.conf
id_type_in_filename = ip
store_slave_file_use_link = false
rotate_error_log = false
error_log_rotate_time=00:00
rotate_error_log_size = 0
use_connection_pool = false
connection_pool_max_idle_time = 3600
http.server_port=8080
http.check_alive_interval=30
http.check_alive_type=tcp
http.check_alive_uri=/status.html

Answer 1

可以发一个源码，参考一下吗？我也遇到这个问题了

Answer 2

带宽是多少呀？

Answer 3

解决了，写了个连接池。

Answer 4

回复
请问，具体如何解决的？谢谢

Answer 5

请问问题得到解决了吗

Answer 6

用common-pool整个链接池吧