Если кто нибудь сталкивался с такой проблемой помогите разобраться
переодическое падение производительности базы.
проблема исчезает после выполнения компрессии
sdeversion -o compress -u sde -p zeta30 -N
Параметры сервера и БД прилагаю ниже.
Нагрузка на сервер 75 пользователей
Win2000 Server SP4 два процессора Intel Xeon 2Gb RAM 5*36Gb HDD SCSI
Oracle 8.1.7 ArcSDE 8.3
Параметры БД
# ================================================
db_name = "dba1"
instance_name = dba1
service_names = dba1
compatible = 8.1.0
remote_login_passwordfile = exclusive
os_authent_prefix = ""
# ================================================
# ================================================
control_files = (
"c:\orabases\dba3\control\control01.ctl",
"d:\orabases\dba3\control\control02.ctl",
"e:\orabases\dba3\control\control03.ctl",
"f:\orabases\dba3\control\control04.ctl",
"g:\orabases\dba3\control\control05.ctl"
)
# ================================================
# ================================================
db_files = 1024
open_cursors = 300
processes = 200
distributed_transactions = 10
max_enabled_roles = 75
timed_statistics = true
# ================================================
# ================================================
job_queue_interval = 3600
job_queue_processes = 1
# ================================================
# ================================================
optimizer_index_cost_adj = 90
optimizer_index_caching = 90
# ================================================
# ================================================
#pre_page_sga = true
db_block_buffers = 15000
#shared_pool_size = 236978176
shared_pool_size = 268435456
log_buffer = 409600
db_block_size = 8192
sort_area_size = 319488
sort_area_retained_size = 319488
db_file_multiblock_read_count = 8
#session_cached_cursors = 25
#shared_pool_reserved_size = 5924454
#PARALLEL_AUTOMATIC_TUNING = true
# ================================================
# ================================================
log_archive_start = true
log_archive_dest_1 = "location=d:\orabases\dba3\arc mandatory reopen=1800"
log_archive_dest_2 = "location=e:\orabases\dba3\arc mandatory reopen=1800"
log_archive_dest_3 = "location=f:\orabases\dba3\arc mandatory reopen=1800"
log_archive_dest_4 = "location=g:\orabases\dba3\arc mandatory reopen=1800"
log_archive_format = %S.arc
#log_archive_max_processes = 2
log_checkpoint_interval = 0
log_checkpoint_timeout = 0
# ================================================
# ================================================
background_dump_dest = c:\orabases\dba3\bdump
user_dump_dest = c:\orabases\dba3\udump
utl_file_dir = (c:\orabases\dba3\udump, c:\rnion)
max_dump_file_size = 5120
# ================================================
кусочек из trc файла
*** 2005-08-12 21:42:48.812
Probe:read_pipe: receive failed, status 1
Probe:S:debug_loop: timeout. Action 1
FATAL ERROR IN TWO-TASK SERVER: error = 12571
*** 2005-08-12 21:42:48.812
ksedmp: internal or fatal error
Current SQL statement for this session:
begin
sys.dbms_debug.default_timeout := 3600;
sys.dbms_debug.debug_off;
end;
----- Call Stack Trace -----
calling   ;   ; call entry & nbsp; & nbsp; argument values in hex
location &nbs p; &nbs p; type point & nbsp; & nbsp; (? means dubious value)
-------------------- &nbs p; -------- -------------------- &nbs p; ----------------------------
_ksedmp+a8 &n bsp; &n bsp; CALLrel _ksedst+0 &nb sp; &nb sp;
FEE47C
_opitsk+f50 & nbsp; & nbsp; CALLrel _ksedmp+0 &nb sp; &nb sp; 2
_opiino+50c & nbsp; & nbsp; CALLrel _opitsk+0 &nb sp; &nb sp; 0
_opiodr+506 & nbsp; & nbsp; CALLreg 00000000 &nbs p; &nbs p; 3C 4
2FEFBF8
_opidrv+384 & nbsp; & nbsp; CALLrel _opiodr+0 &nb sp; &nb sp; 3C 4
2FEFBF8
0
_sou2o+19 
