Little Server lags

[UnAfraid]

» Find Revision
L2J Revision 4175:
L2JDP Revision 7405:

I've got little problems with lags its not from internet is not from pc ..
The server is moving slow some of my actions do not have effect

There is no errors in log

Server Machine Configuration:

• CPU: 2x Intel(R) Core(TM) i7 CPU 975 @ 3.33GHz
  RAM: 12 GB DDR3
  HDD: 3 TB / Software raid 5
  OS: Centos x64

Startup Parameters:

• java -Xmx8192m -cp ./../libs/*:l2jserver.jar com.l2jserver.gameserver.GameServer > log/stdout.log 2>&1

Etc settings:

• ulimit -n 8192

(i am using this because of error "Too many open files")

MySQL Config:

Code: Select all

 [client]#password       = [your_password]port            = 3306socket          = /var/run/mysqld/mysqld.sock [mysqld]user            = mysqlpid-file        = /var/run/mysqld/mysqld.pidsocket          = /var/run/mysqld/mysqld.sockport            = 3306basedir         = /usrdatadir         = /var/lib/mysqltmpdir          = /tmplanguage        = /usr/share/mysql/englishdefault-storage-engine=INNODBback_log = 50#skip-networkingmax_connections = 600max_connect_errors = 0table_cache = 256#external-lockingmax_allowed_packet = 16Mbinlog_cache_size = 1Mmax_heap_table_size = 64Msort_buffer_size = 8Mjoin_buffer_size = 8Mthread_cache_size = 8thread_concurrency = 8#query_cache_size = 64Mquery_cache_limit = 2Mft_min_word_len = 4default_table_type = MYISAM#memlockthread_stack = 192Ktransaction_isolation = REPEATABLE-READtmp_table_size = 256M#log_slave_updates#log#log_warningslong_query_time = 2#tmpdir = /tmpserver-id = 1   #*** MyISAM Specific optionskey_buffer_size = 32Mread_buffer_size = 2Mread_rnd_buffer_size = 16Mbulk_insert_buffer_size = 64Mmyisam_sort_buffer_size = 69Mmyisam_max_sort_file_size = 100Gmyisam_max_extra_sort_file_size = 10Gmyisam_repair_threads = 1myisam_recover  # *** BDB Specific options ***#skip-bdb  # *** INNODB Specific options ***#skip-innodbinnodb_additional_mem_pool_size = 8Minnodb_buffer_pool_size = 1Ginnodb_data_file_path = ibdata1:10M:autoextend#innodb_data_home_dir = <directory>innodb_file_io_threads = 4#innodb_force_recovery=1innodb_thread_concurrency = 60innodb_flush_log_at_trx_commit = 2#innodb_fast_shutdowninnodb_log_buffer_size = 16Minnodb_log_file_size = 512Minnodb_log_files_in_group = 3#innodb_log_group_home_dirinnodb_max_dirty_pages_pct = 90#innodb_flush_method=O_DSYNCinnodb_lock_wait_timeout = 120innodb_file_per_table  [mysqldump]quick max_allowed_packet = 16M [mysql]no-auto-rehash # Only allow UPDATEs and DELETEs that use keys.#safe-updates [isamchk]key_buffer = 512Msort_buffer_size = 512Mread_buffer = 8Mwrite_buffer = 8M [myisamchk]key_buffer = 512Msort_buffer_size = 512Mread_buffer = 8Mwrite_buffer = 8M [mysqlhotcopy]interactive-timeout [mysqld_safe]open-files-limit = 8192  

Thread configuration:

Code: Select all

 # ---------------------------------------------------------------------------# Thread Configuration# ---------------------------------------------------------------------------# Extreme caution should be here, set to defaults if you do not know what you are doing.# These could possibly hurt your servers performance or improve it depending on your server's configuration, size, and other factors.# Default: 10ThreadPoolSizeEffects = 30 # Default: 13ThreadPoolSizeGeneral = 39 # Default: 2UrgentPacketThreadCoreSize = 4 # Default: 4GeneralPacketThreadCoreSize = 8 # Default: 4GeneralThreadCoreSize = 8 # Default: 6AiMaxThread = 12 # Dead Lock Detector (a separate thread for detecting deadlocks).# For improved crash logs and automatic restart in deadlock case if enabled.# Check interval is in seconds.# Default: FalseDeadLockDetector = False # Default: 20DeadLockCheckInterval = 20 # Default: FalseRestartOnDeadlock = False 

Is there any problems in my configuration ?

[MELERIX]

try with "Thread Configuration" has default

[UnAfraid]

I will thanks, Btw can i receive information about them what they actualy do?

[blacksea]

UnAfraid did you find the reason why you get "Too many open files" ?
i have same problem and i can`t find where is the problem , ulimit -n will fix it? i didn`t try yet.

[JIV]

your starting script blows.

[blacksea]

your starting script blows.

but i didn't have problems until i changed server

[Gnacik]

For sure you should try -server param to java. also default java gc just suxx

[Nik]

Lol... cool startup script

It should look smth like this

Code: Select all

nice -n -3 java -server -Xmx6128m -Xms3092m -Xmn1024m -Xss128k -XX:PermSize=128m -XX:+UseThreadPriorities -XX:ParallelGCThreads=2 -XX:+UseConcMarkSweepGC -XX:+UseParNewGC -XX:SurvivorRatio=8 -XX:TargetSurvivorRatio=90 -Xnoclassgc -XX:+AggressiveOpts -Djava.util.logging.manager=com.l2jserver.util.L2LogManager -cp ./../libs/****-1.9.5.jar:./../libs/*:l2jserver.jar com.****.**** com.l2jserver.gameserver.GameServer > log/stdout.log 2>&1

[JIV]

afaik default thread stack size at 64b jvm is 512kb

[blacksea]

current:

Code: Select all

java -server -Djava.util.logging.manager=com.l2jserver.util.L2LogManager -Xmx12G -Xms12G -Xmn3G -XX:PermSize=256m -XX:SurvivorRatio=16 -Xnoclassgc -XX:+AggressiveOpts -cp ./../libs/*:l2jserver.jar com.l2jserver.gameserver.GameServer 

i`ll try nik suggestion .. but Xmx and Xms should be same from what i documented.

[jurchiks]

[OFFTOPIC]@nik - somebody has problems with math it seems:
-Xmx6128m -Xms3092m
should be
-Xmx6144m -Xms3072m
Or even better
-Xmx6g -Xms3g