oninit process consumes memory

We are using Informix Dynamic Server 9.4 on HPUX11i. We have observed = that the memory consumed by oninit process steadily increases over a = period of time. We find that there is no apparent memory leakage in our = application that processes data in the database. Even if we shutdown all = the application processes, the oninit process does not release the = memory. Executing 'onmode -F' also does not help. =20 Most of our applications are written in ESQL/C and Java1.4 language. The = Java uses JDBC 2.0 connection whereas ESQL/C uses shared memory to = connect to database. =20 The HPUX11i is a 3CPU rp5470 HP-UX box with 3 GB RAM with 10 GB Swap = Space. =20 Can anybody give any hint as why the oninit memory is increasing = foreever and how to force it to release it. =20 Following is the output of top command on HP-UX with oninit process = highlighted. =20 #########################################################################= # System: rp5470 Mon Feb 23 16:11:36 2004 Load averages: 0.35, 0.37, 0.38 270 processes: 250 sleeping, 20 running Cpu states: CPU LOAD USER NICE SYS IDLE BLOCK SWAIT INTR SSYS 0 0.62 56.9% 0.0% 0.6% 42.5% 0.0% 0.0% 0.0% 0.0%=20 1 0.18 24.0% 0.0% 0.4% 75.6% 0.0% 0.0% 0.0% 0.0%=20 2 0.26 31.9% 0.0% 13.8% 54.3% 0.0% 0.0% 0.0% 0.0%=20 --- ---- ----- ----- ----- ----- ----- ----- ----- ----- avg 0.35 37.6% 0.0% 4.8% 57.6% 0.0% 0.0% 0.0% 0.0%=20 Memory: 1517120K (1361396K) real, 3105448K (2502008K) virtual, 75520K = free Page# 1/1 CPU TTY PID USERNAME PRI NI SIZE RES STATE TIME %WCPU %CPU COMMAND 0 ? 24470 informix 180 20 1275M 576K run 6245:42 52.95 52.86 oninit 2 ? 24472 informix 180 20 1275M 576K sleep 8078:08 47.53 47.45 oninit 2 ? 28173 root 152 20 267M 75864K run 254:05 3.43 3.42 java 0 pts/tc 404 root 152 20 314M 125M run 1838:53 2.70 2.69 java 0 pts/tc 12059 root 152 20 245M 78628K run 185:06 1.33 1.33 java 1 ? 2377 root 152 24 263M 45172K run 128:16 0.60 0.60 rmiregistry 2 pts/tj 12375 root 154 10 28948K 4368K sleep 71:47 0.41 0.41 gpm 0 ? 23783 root 152 20 357M 5116K run 5:53 0.39 0.39 slapd 2 ? 36 root 152 20 7264K 7264K run 59:46 0.37 0.37 vxfsd 1 ? 14298 informix 152 20 212M 5016K run 121:54 0.31 0.30 java 2 ? 1650 root 152 20 214M 4476K run 119:03 0.30 0.30 prm3d 2 ? 16441 root -16 20 23864K 8528K run 114:21 0.24 0.24 midaemon 1 ? 1400 root 152 20 14476K 1832K run 0:50 0.23 0.23 dmisp 1 ? 3085 root 152 20 29364K 13512K run 13:08 0.23 0.23 rds 1 ? 8309 root 154 20 3712K 268K sleep 96:57 0.16 0.16 nmbd 2 ? 24478 informix 180 20 1275M 160K sleep 59:24 0.13 0.13 oninit #########################################################################= # =20 Following is the extract from onconfig file for INFORMIX SERVER =20 # Shared Memory Parameters =20 LOCKS 100000 # Maximum number of locks BUFFERS 375000 # Maximum number of shared buffers NUMAIOVPS 2 # Number of IO vps PHYSBUFF 32 # Physical log buffer size (Kbytes) LOGBUFF 64 # Logical log buffer size (Kbytes) CLEANERS 2 # Number of buffer cleaner processes SHMBASE 0x0 # Shared memory base address SHMVIRTSIZE 350000 # initial virtual shared memory segment =size SHMADD 16384 # Size of new shared memory segments = (Kbytes) SHMTOTAL 0 # Total shared memory (Kbytes). = 0=3D>unlimited CKPTINTVL 300 # Check point interval (in sec) LRUS 8 # Number of LRU queues LRU_MAX_DIRTY 60.000000 # LRU percent dirty begin cleaning limit LRU_MIN_DIRTY 50.000000 # LRU percent dirty end cleaning limit TXTIMEOUT 0x12c # Transaction timeout (in sec) STACKSIZE 64 # Stack size (Kbytes) #########################################################################= # Regards,=20 Rohit=20 =20

run onstat -g stmIf there is allot of open cursors your application needs to do a better job of handling them. Also why is RESIDENT not set? I not sure about HP but with Solaris you can get in some real memory issues if it it not. On Wed, 2004-02-25 at 06:36, Kerhalkar, .... wrote: > We are using Informix Dynamic Server 9.4 on HPUX11i. We have observed = > that the memory consumed by oninit process steadily increases over a = > period of time. We find that there is no apparent memory leakage in our = > application that processes data in the database. Even if we shutdown all = > the application processes, the oninit process does not release the = > memory. Executing 'onmode -F' also does not help. > =20 > Most of our applications are written in ESQL/C and Java1.4 language. The = > Java uses JDBC 2.0 connection whereas ESQL/C uses shared memory to = > connect to database. > =20 > The HPUX11i is a 3CPU rp5470 HP-UX box with 3 GB RAM with 10 GB Swap = > Space. > =20 > Can anybody give any hint as why the oninit memory is increasing = > foreever and how to force it to release it. > =20 > Following is the output of top command on HP-UX with oninit process = > highlighted. > =20 > #########################################################################= > # > System: rp5470 Mon Feb 23 16:11:36 2004 > > Load averages: 0.35, 0.37, 0.38 > > 270 processes: 250 sleeping, 20 running > > Cpu states: > > CPU LOAD USER NICE SYS IDLE BLOCK SWAIT INTR SSYS > > 0 0.62 56.9% 0.0% 0.6% 42.5% 0.0% 0.0% 0.0% 0.0%=20 > > 1 0.18 24.0% 0.0% 0.4% 75.6% 0.0% 0.0% 0.0% 0.0%=20 > > 2 0.26 31.9% 0.0% 13.8% 54.3% 0.0% 0.0% 0.0% 0.0%=20 > > --- ---- ----- ----- ----- ----- ----- ----- ----- ----- > > avg 0.35 37.6% 0.0% 4.8% 57.6% 0.0% 0.0% 0.0% 0.0%=20 > > Memory: 1517120K (1361396K) real, 3105448K (2502008K) virtual, 75520K = > free Page# 1/1 > > CPU TTY PID USERNAME PRI NI SIZE RES STATE TIME %WCPU %CPU COMMAND > > 0 ? 24470 informix 180 20 1275M 576K run 6245:42 52.95 52.86 oninit > > 2 ? 24472 informix 180 20 1275M 576K sleep 8078:08 47.53 47.45 oninit > > 2 ? 28173 root 152 20 267M 75864K run 254:05 3.43 3.42 java > > 0 pts/tc 404 root 152 20 314M 125M run 1838:53 2.70 2.69 java > > 0 pts/tc 12059 root 152 20 245M 78628K run 185:06 1.33 1.33 java > > 1 ? 2377 root 152 24 263M 45172K run 128:16 0.60 0.60 rmiregistry > > 2 pts/tj 12375 root 154 10 28948K 4368K sleep 71:47 0.41 0.41 gpm > > 0 ? 23783 root 152 20 357M 5116K run 5:53 0.39 0.39 slapd > > 2 ? 36 root 152 20 7264K 7264K run 59:46 0.37 0.37 vxfsd > > 1 ? 14298 informix 152 20 212M 5016K run 121:54 0.31 0.30 java > > 2 ? 1650 root 152 20 214M 4476K run 119:03 0.30 0.30 prm3d > > 2 ? 16441 root -16 20 23864K 8528K run 114:21 0.24 0.24 midaemon > > 1 ? 1400 root 152 20 14476K 1832K run 0:50 0.23 0.23 dmisp > > 1 ? 3085 root 152 20 29364K 13512K run 13:08 0.23 0.23 rds > > 1 ? 8309 root 154 20 3712K 268K sleep 96:57 0.16 0.16 nmbd > > 2 ? 24478 informix 180 20 1275M 160K sleep 59:24 0.13 0.13 oninit > > #########################################################################= > # > =20 > Following is the extract from onconfig file for INFORMIX SERVER > =20 > # Shared Memory Parameters > =20 > LOCKS 100000 # Maximum number of locks > BUFFERS 375000 # Maximum number of shared buffers > NUMAIOVPS 2 # Number of IO vps > PHYSBUFF 32 # Physical log buffer size (Kbytes) > LOGBUFF 64 # Logical log buffer size (Kbytes) > CLEANERS 2 # Number of buffer cleaner processes > SHMBASE 0x0 # Shared memory base address > SHMVIRTSIZE 350000 # initial virtual shared memory segment => size > SHMADD 16384 # Size of new shared memory segments = > (Kbytes) > SHMTOTAL 0 # Total shared memory (Kbytes). = > 0=3D>unlimited > CKPTINTVL 300 # Check point interval (in sec) > LRUS 8 # Number of LRU queues > LRU_MAX_DIRTY 60.000000 # LRU percent dirty begin cleaning limit > LRU_MIN_DIRTY 50.000000 # LRU percent dirty end cleaning limit > TXTIMEOUT 0x12c # Transaction timeout (in sec) > STACKSIZE 64 # Stack size (Kbytes) > #########################################################################= > #> > Regards,=20 > Rohit=20 > > =20 > > >

Hi, reasons for this behaviour can be manifold, as memory is used for many different purposes in the engine ... What you should do is to provide some more info for analysis. To aid analysis you can execute the following statements which will retrieve much more detailed info about memory usage by the engine: onstat -g seg onstat -g mem onstat -g mgm onstat -g ufr <pool name> # where pool name can be retrieved from above"onstat -g mem" You can also run above commands periodically to (hopefully) better show, in which area the memory usage increases ... For ideas on further commands please see the usage message of onstat (e.g. do "onstat --"). TIA, Martin -- Martin Fuerderer IBM Informix Development Munich Data Management Solutions "Kerhalkar, ...." <rohit.kerhalkar@hp.com> Sent by: forum.subscriber@iiug.org 25.02.2004 12:36 To ids@iiug.org cc Subject oninit process consumes memory [2563] We are using Informix Dynamic Server 9.4 on HPUX11i. We have observed = that the memory consumed by oninit process steadily increases over a = period of time. We find that there is no apparent memory leakage in our = application that processes data in the database. Even if we shutdown all = the application processes, the oninit process does not release the = memory. Executing 'onmode -F' also does not help. =20 Most of our applications are written in ESQL/C and Java1.4 language. The = Java uses JDBC 2.0 connection whereas ESQL/C uses shared memory to = connect to database. =20 The HPUX11i is a 3CPU rp5470 HP-UX box with 3 GB RAM with 10 GB Swap = Space. =20 Can anybody give any hint as why the oninit memory is increasing = foreever and how to force it to release it. =20 Following is the output of top command on HP-UX with oninit process = highlighted. =20 #########################################################################= # System: rp5470 Mon Feb 23 16:11:36 2004 Load averages: 0.35, 0.37, 0.38 270 processes: 250 sleeping, 20 running Cpu states: CPU LOAD USER NICE SYS IDLE BLOCK SWAIT INTR SSYS 0 0.62 56.9% 0.0% 0.6% 42.5% 0.0% 0.0% 0.0% 0.0%=20 1 0.18 24.0% 0.0% 0.4% 75.6% 0.0% 0.0% 0.0% 0.0%=20 2 0.26 31.9% 0.0% 13.8% 54.3% 0.0% 0.0% 0.0% 0.0%=20 --- ---- ----- ----- ----- ----- ----- ----- ----- ----- avg 0.35 37.6% 0.0% 4.8% 57.6% 0.0% 0.0% 0.0% 0.0%=20 Memory: 1517120K (1361396K) real, 3105448K (2502008K) virtual, 75520K = free Page# 1/1 CPU TTY PID USERNAME PRI NI SIZE RES STATE TIME %WCPU %CPU COMMAND 0 ? 24470 informix 180 20 1275M 576K run 6245:42 52.95 52.86 oninit 2 ? 24472 informix 180 20 1275M 576K sleep 8078:08 47.53 47.45 oninit 2 ? 28173 root 152 20 267M 75864K run 254:05 3.43 3.42 java 0 pts/tc 404 root 152 20 314M 125M run 1838:53 2.70 2.69 java 0 pts/tc 12059 root 152 20 245M 78628K run 185:06 1.33 1.33 java 1 ? 2377 root 152 24 263M 45172K run 128:16 0.60 0.60 rmiregistry 2 pts/tj 12375 root 154 10 28948K 4368K sleep 71:47 0.41 0.41 gpm 0 ? 23783 root 152 20 357M 5116K run 5:53 0.39 0.39 slapd 2 ? 36 root 152 20 7264K 7264K run 59:46 0.37 0.37 vxfsd 1 ? 14298 informix 152 20 212M 5016K run 121:54 0.31 0.30 java 2 ? 1650 root 152 20 214M 4476K run 119:03 0.30 0.30 prm3d 2 ? 16441 root -16 20 23864K 8528K run 114:21 0.24 0.24 midaemon 1 ? 1400 root 152 20 14476K 1832K run 0:50 0.23 0.23 dmisp 1 ? 3085 root 152 20 29364K 13512K run 13:08 0.23 0.23 rds 1 ? 8309 root 154 20 3712K 268K sleep 96:57 0.16 0.16 nmbd 2 ? 24478 informix 180 20 1275M 160K sleep 59:24 0.13 0.13 oninit #########################################################################= # =20 Following is the extract from onconfig file for INFORMIX SERVER =20 # Shared Memory Parameters =20 LOCKS 100000 # Maximum number of locks BUFFERS 375000 # Maximum number of shared buffers NUMAIOVPS 2 # Number of IO vps PHYSBUFF 32 # Physical log buffer size (Kbytes) LOGBUFF 64 # Logical log buffer size (Kbytes) CLEANERS 2 # Number of buffer cleaner processes SHMBASE 0x0 # Shared memory base address SHMVIRTSIZE 350000 # initial virtual shared memory segment =size SHMADD 16384 # Size of new shared memory segments = (Kbytes) SHMTOTAL 0 # Total shared memory (Kbytes). = 0=3D>unlimited CKPTINTVL 300 # Check point interval (in sec) LRUS 8 # Number of LRU queues LRU_MAX_DIRTY 60.000000 # LRU percent dirty begin cleaning limit LRU_MIN_DIRTY 50.000000 # LRU percent dirty end cleaning limit TXTIMEOUT 0x12c # Transaction timeout (in sec) STACKSIZE 64 # Stack size (Kbytes) #########################################################################= # Regards,=20 Rohit=20 =20