Update from Digital 7.31.FD1to ....
After upgrading IDS from 7.31.FD1 to FD7 on a Tru64/Alpha ES40 (4 CPUs, 4GB) running Baan IV, overnight batch jobs got faster but interactive performance was poor once ~150 users were online; the poster asked whether ONCONFIG tuning carried over from FD1 needed changing. Suggestions included raising LRUS/CLEANERS, checking KAIO (onstat -g ioq confirmed kio queues active), re-running UPDATE STATISTICS after the upgrade, and NETTYPE changes — Ralph Gentry advised moving soctcp poll threads to CPU VPs, which Art Kagel disputed, arguing TCP poll threads should stay in NET VPs. No resolution is recorded.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
7.31.FD7.
Hello!
After endless problems
(Thread http://www.iiug.org/forums/ids/index.cgi/noframes/read/9445)
i make now with this Digital UNIX V4.0F version tha last possible update to
version 7.31.FD7
i change all kernelparameters what is need and stand in the releasedocuments.
yesterday i must give the machine more swap space because the operator befor
not give double mainmemory and it seems to me that this version from informix
is now able to use the hole 4GB memory in the machine. (at moment i give the
machine around 2 GB)
after restart and many jobs (baan iv) in the night i was happy because this
jobs was much more early ready than with the old informix version 7.31FD1.
.............but................ today now when more than 150 peoples online
the performance from this machine is not good.
so now my question to all specialists.
form version FD1 to FD7 are many fixes and want to know if i must make other
onconfig parameters.
the onconfig parameters from FD1 was long long time tuned to best performance.
onstat -p
Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 20:37:26 -- 2039808
Kbytes
Profile
dskreads pagreads bufreads %cached dskwrits pagwrits bufwrits %cached
18261777 68699403 8147783805 99.78 1112070 14149844 4388198 74.66
isamtot open start read write rewrite delete commit rollbk
15732299654 10327635 74498653 7379610531 460406 187719 184098 208520 17
gp_read gp_write gp_rewrt gp_del gp_alloc gp_free gp_curs
0 0 0 0 0 0 0
ovlock ovuserthread ovbuff usercpu syscpu numckpts flushes
0 0 0 107920.02 3463.92 124 248
bufwaits lokwaits lockreqs deadlks dltouts ckpwaits compress seqscans
2683063 17 240204811 0 0 210 60672 232903
ixda-RA idx-RA da-RA RA-pgsused lchwaits
5142782 4392090 1502752 10718689 608758
Robert
At the Unix level, what is causing the performance issue?
What changes have you made to the Informix config?
Are you getting more sequential scans than previous? Guessing from your
stats, probably not.
You need to determine what the basic issue is with the server or
application. Some people here could make suggestions to your ONCONFIG
is we know more about your problem.
MW
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
ROBERT LANGHAMMER
Sent: Wednesday, 28 November 2007 3:49 a.m.
To: ids@iiug.org
Subject: Update from Digital 7.31.FD1to .... [10484]
7.31.FD7.
Hello!
After endless problems
(Thread http://www.iiug.org/forums/ids/index.cgi/noframes/read/9445)
i make now with this Digital UNIX V4.0F version tha last possible update
to
version 7.31.FD7
i change all kernelparameters what is need and stand in the
releasedocuments.
yesterday i must give the machine more swap space because the operator
befor
not give double mainmemory and it seems to me that this version from
informix
is now able to use the hole 4GB memory in the machine. (at moment i give
the
machine around 2 GB)
after restart and many jobs (baan iv) in the night i was happy because
this
jobs was much more early ready than with the old informix version
7.31FD1.
..............but................ today now when more than 150 peoples
online
the performance from this machine is not good.
so now my question to all specialists.
form version FD1 to FD7 are many fixes and want to know if i must make
other
onconfig parameters.
the onconfig parameters from FD1 was long long time tuned to best
performance.
onstat -p
Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 20:37:26 --
2039808
Kbytes
Profile
dskreads pagreads bufreads %cached dskwrits pagwrits bufwrits %cached
18261777 68699403 8147783805 99.78 1112070 14149844 4388198 74.66
isamtot open start read write rewrite delete commit rollbk
15732299654 10327635 74498653 7379610531 460406 187719 184098 208520 17
gp_read gp_write gp_rewrt gp_del gp_alloc gp_free gp_curs
0 0 0 0 0 0 0
ovlock ovuserthread ovbuff usercpu syscpu numckpts flushes
0 0 0 107920.02 3463.92 124 248
bufwaits lokwaits lockreqs deadlks dltouts ckpwaits compress seqscans
2683063 17 240204811 0 0 210 60672 232903
ixda-RA idx-RA da-RA RA-pgsused lchwaits
5142782 4392090 1502752 10718689 608758
************************************************************************
*******
Forum Note: Use "Reply" to post a response in the discussion forum.
Okay here the machine data and onconfig:
The machine is a Compaq True 64 Unix V4.0 1229 alpha with 4 CPUs and 4 GB
mainmemory.
The only changes what we make is what stand in the releasenotes:
shmmax = not change (2147483647)
shmmni from 512 to 2048
shmseg from 150 to 512
semmni 400 (not changed)
semmsl 1600 (not changed)
semopm 400 (not changed)
semume 400 (not changed)
maxdsiz ( max-per-proc-data-size ) should change to 9223372036854775807 but
kernel let max 4398046511104
maxvas (vm-maxvas) from 2048000000 to 9223372036854775807
per-proc-adress-space (per-proc-address-space) from 2147483647 to 4398046511104
max-per-proc-address-space (max-per-proc-address-space) from 2147483647 to
4398046511104
onconfig are the same than before (temporary we change SHMVIRTSIZE from
1100000 to 3317760 but the admin befor make to smal swap space and we get
problems so i turn this parameter back. now i put one more disk inside and
make now biger swap space. (more than double mainmemory)
onconfig:
#**************************************************************************
#
# INFORMIX SOFTWARE, INC.
#
# Title: onconfig.baan
# Description: INFORMIX-OnLine Configuration Parameters
#
#**************************************************************************
# Root Dbspace Configuration
ROOTNAME rootdbs # Root dbspace name
ROOTPATH /dev/syschunk # Path for device containing root dbspace
ROOTOFFSET 0 # Offset of root dbspace into device (Kbytes)
ROOTSIZE 40000 # Size of root dbspace (Kbytes)
# Disk Mirroring Configuration Parameters
MIRROR 0 # Mirroring flag (Yes = 1, No = 0)
MIRRORPATH # Path for device containing mirrored root
MIRROROFFSET 0 # Offset into mirrored device (Kbytes)
# Physical Log Configuration
PHYSDBS plogdbs # Location (dbspace) of physical log
PHYSFILE 299000 # Physical log file size (Kbytes)
# Logical Log Configuration
LOGFILES 100 # Number of logical log files
LOGSIZE 10000 # Logical log size (Kbytes)
# Diagnostics
MSGPATH /informix/online.log # System message log file path
CONSOLE /dev/console # System console message path
ALARMPROGRAM /informix/etc/log_full.sh # Alarm program path
#ALARMPROGRAM /informix/etc/no_log.sh # Alarm program path
TBLSPACE_STATS 1
# System Archive Tape Device
TAPEDEV /dev/null # Tape device path
TAPEBLK 64 # Tape block size (Kbytes)
TAPESIZE 61440000 # Maximum amount of data to put on tape (Kbytes)
# Log Archive Tape Device
LTAPEDEV /tmp/log.log # Log tape device path
#LTAPEDEV /dev/null # Log tape device path
LTAPEBLK 64 # Log tape block size (Kbytes)
LTAPESIZE 61440000 # Max amount of data to put on log tape (Kbytes)
# Optical
STAGEBLOB # INFORMIX-OnLine/Optical staging area
# System Configuration
SERVERNUM 1 # Unique id corresponding to a OnLine instance
DBSERVERNAME inf_baan # Name of default database server
DBSERVERALIASES inf_normal # List of alternate dbservernames
NETTYPE soctcp,8,50,NET # Configure poll thread(s) for nettype
NETTYPE ipcstr,1,60,CPU # Configure poll thread(s) for nettype
NETTYPE sqlmux # Configure poll thread(s) for nettype
DEADLOCK_TIMEOUT 60 # Max time to wait of lock in distributed env.
RESIDENT 0 # Forced residency flag (Yes = 1, No = 0)
MULTIPROCESSOR 1 # 0 for single-processor, 1 for multi-processor
NUMCPUVPS 3 # Number of user (cpu) vps
SINGLE_CPU_VP 0 # If non-zero, limit number of cpu vps to one
NOAGE 1 # Process aging
AFF_SPROC 1 # Affinity start processor
AFF_NPROCS 3 # Affinity number of processors
# Shared Memory Parameters
LOCKS 500000 # Maximum number of locks
BUFFERS 300000#BUFFERS 200000 # 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)LOGSMAX 100 # Maximum number of logical log files
CLEANERS 8 # Number of buffer cleaner processes
SHMBASE 0x200000000 # Shared memory base address
SHMVIRTSIZE 1100000#SHMVIRTSIZE 753664 # initial virtual shared memory segment size
SHMADD 57344 # Size of new shared memory segments (Kbytes)
SHMTOTAL 0 # Total shared memory (Kbytes). 0=>unlimited
CKPTINTVL 600 # Check point interval (in sec)
LRUS 8 # Number of LRU queues
LRU_MAX_DIRTY 5 # LRU percent dirty begin cleaning limit
LRU_MIN_DIRTY 2 # LRU percent dirty end cleaning limit
LTXHWM 50 # Long transaction high water mark percentage
LTXEHWM 60 # Long transaction high water mark (exclusive)
TXTIMEOUT 0x12c # Transaction timeout (in sec)
STACKSIZE 64 # Stack size (Kbytes)
# System Page Size
# BUFFSIZE - OnLine no longer supports this configuration parameter.
# To determine the page size used by OnLine on your platform
# see the last line of output from the command, 'onstat -b'.
# Recovery Variables
# OFF_RECVRY_THREADS:
# Number of parallel worker threads during fast recovery or an offline restore.
# ON_RECVRY_THREADS:
# Number of parallel worker threads during an online restore.
OFF_RECVRY_THREADS 10 # Default number of offline worker threads
ON_RECVRY_THREADS 1 # Default number of online worker threads
# Data Replication Variables
# DRAUTO: 0 manual, 1 retain type, 2 reverse type
DRAUTO 0 # DR automatic switchover
DRINTERVAL 30 # DR max time between DR buffer flushes (in sec)
DRTIMEOUT 30 # DR network timeout (in sec)DRLOSTFOUND /informix/etc/dr.lostfound # DR lost+found file path
# CDR Variables
CDR_LOGBUFFERS 2048 # size of log reading buffer pool (Kbytes)
CDR_EVALTHREADS 1,2 # evaluator threads (per-cpu-vp,additional)
CDR_DSLOCKWAIT 5 # DS lockwait timeout (seconds)
CDR_QUEUEMEM 4096 # Maximum amount of memory for any CDR queue (Kbytes)
# Backup/Restore variables
BAR_BSALIB_PATH /usr/lib/libbsa.so
BAR_ACT_LOG /informix/bar_act.log
BAR_MAX_BACKUP 1
BAR_RETRY 1
BAR_NB_XPORT_COUNT 10
BAR_XFER_BUF_SIZE 31
ISM_DATA_POOL RocoDaten
ISM_LOG_POOL RocoLogs
# Read Ahead Variables
RA_PAGES 72 # Number of pages to attempt to read ahead
RA_THRESHOLD 66 # Number of pages left before next group
# DBSPACETEMP:
# OnLine equivalent of DBTEMP for SE. This is the list of dbspaces
# that the OnLine SQL Engine will use to create temp tables etc.
# If specified it must be a colon separated list of dbspaces that exist
# when the OnLine system is brought online. If not specified, or if
# all dbspaces specified are invalid, various ad hoc queries will create
# temporary files in /tmp instead.
DBSPACETEMP tempdbs,tempdbs1 # Default temp dbspaces
# DUMP*:
# The following parameters control the type of diagnostics information which
# is preserved when an unanticipated error condition (assertion failure) occurs
# during OnLine operations.
# For DUMPSHMEM, DUMPGCORE and DUMPCORE 1 means Yes, 0 means No.
DUMPDIR /tmp # Preserve diagnostics in this directory
DUMPSHMEM 0 # Dump a copy of shared memory
DUMPGCORE 0 # Dump a core image using 'gcore'
DUMPCORE 0 # Dump a core image (Warning:this aborts OnLine)
DUMPCNT 1 # Number of shared memory or gcore dumps for
# a single user's session
FILLFACTOR 90 # Fill factor for building indexes
@@NL@
sysconfigtab: # # ***************************************************************** # * * # * Copyright (c) Digital Equipment Corporation, 1991, 1999 * # * * # * All Rights Reserved. Unpublished rights reserved under * # * the copyright laws of the United States. * # * * # * The software contained on this media is proprietary to * # * and embodies the confidential technology of Digital * # * Equipment Corporation. Possession, use, duplication or * # * dissemination of the software and media is authorized only * # * pursuant to a valid written license from Digital Equipment * # * Corporation. * # * * # * RESTRICTED RIGHTS LEGEND Use, duplication, or disclosure * # * by the U.S. Government is subject to restrictions as set * # * forth in Subparagraph (c)(1)(ii) of DFARS 252.227-7013, * # * or in FAR 52.227-19, as applicable. * # * * # ***************************************************************** # # HISTORY # # (c) Copyright 1990, 1991, 1992, 1993 OPEN SOFTWARE FOUNDATION, INC. # ALL RIGHTS RESERVED # # # OSF/1 1.2 # # # The supported method of changing the information in this # file (sysconfigtab) is by using the sysconfigdb command. # # If you insist upon changing this file by direct editing, # then it would be advisable to add new information at the end # rather than the beginning or middle of the file, in order # to reduce confusion during installation and reconciliation # of user changes and update changes. # # See the appropriate documentation such as the release notes # for the new version being installed as well as the appropriate # reference pages. # # # Contiguous Memory Allocation for loadable driver subsystems # cma_dd: # # EISA_Option = Board_Id Function_Name Driver_Name Type # eisa: EISA_Option = Board_Id - ADP0001, Function_Name - AHA1740, Driver_Name - aha, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - ADP0002, Function_Name - AHA1740, Driver_Name - aha, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC2500, Function_Name - Null, Driver_Name - envram, Type - C, Adpt_Config - N EISA_Option = Board_Id - ISA1010, Function_Name - 'COM,1' Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - ISA1010, Function_Name - 'COM,2', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - ISA1010, Function_Name - PAR, Driver_Name - lp, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC2A01, Function_Name - SYSMEM, Driver_Name - Null, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC2A01, Function_Name - 'ACECOM,1', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC2A01, Function_Name - 'ACECOM,2', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC2A01, Function_Name - PAR, Driver_Name - lp, Type - C, Adpt_Config - N EISA_Option = Board_Id - MLX0070, Function_Name - Null, Driver_Name - xcr, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - MLX0075, Function_Name - Null, Driver_Name - xcr, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - MLX0077, Function_Name - Null, Driver_Name - xcr, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5000, Function_Name - SYSMEM, Driver_Name - Null, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5000, Function_Name - 'ACECOM,1', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5000, Function_Name - 'ACECOM,2', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5000, Function_Name - PAR, Driver_Name - lp, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5100, Function_Name - SYSMEM, Driver_Name - Null, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5100, Function_Name - 'ACECOM,1', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5100, Function_Name - 'ACECOM,2', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5100, Function_Name - PAR, Driver_Name - lp, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5301, Function_Name - SYSMEM, Driver_Name - Null, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5301, Function_Name - 'ACECOM,1', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5301, Function_Name - 'ACECOM,2', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5301, Function_Name - PAR, Driver_Name - lp, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC6000, Function_Name - 'SYSMEM', Driver_Name - Null, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - ADP0001, Function_Name - AHA1740, Driver_Name - aha, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - ADP0002, Function_Name - AHA1740, Driver_Name - aha, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC2500, Function_Name - Null, Driver_Name - envram, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC2A01, Function_Name - 'ACECOM,1', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC2A01, Function_Name - 'ACECOM,2', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC2A01, Function_Name - PAR, Driver_Name - lp, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC2A01, Function_Name - SYSMEM, Driver_Name - Null, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5000, Function_Name - 'ACECOM,1', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5000, Function_Name - 'ACECOM,2', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5000, Function_Name - PAR, Driver_Name - lp, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5000, Function_Name - SYSMEM, Driver_Name - Null, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5100, Function_Name - 'ACECOM,1', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5100, Function_Name - 'ACECOM,2', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5100, Function_Name - PAR, Driver_Name - lp, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5100, Function_Name - SYSMEM, Driver_Name - Null, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5301, Function_Name - 'ACECOM,1', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5301, Function_Name - 'ACECOM,2', Driver_Name - ace, Int_Aft_Attach - 1, Type - C, Adpt_Config - N EISA_Option = Board_Id - DEC5301, Function_Name - PAR, Driver_N
I'm sure others will have good suggestions but here's mine.
Increase LRU/CLEANERS significantly, like maybe 64 or more
noticed NETTYPE. Put your most used access method on CPU and the other on NET
noticed NETTYPE For soctcp was 8,50. suggest trying 3,125
There are various onstat commands to examine which threads are the most
active. If you not using KAIO you may want to up AIO threads.
Yes, make sure you have KAIO turned on. I know at least at one time the
DIGITAL servers did have the ability to use KAIO.
After the upgrade, did you drop and re-execute UPDATE STATISTICS as
recommended on the Informix webpage:
http://www-1.ibm.com/support/docview.wss?uid=swg21137764
Take care.
Clifton M. Bean
Informix DBA / AIX System Admin
Currency Technics & Metrics
Phone: (972) 812-1411 x244
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of RALPH
GENTRY
Sent: Wednesday, November 28, 2007 8:05 AM
To: ids@iiug.org
Subject: Re: Update from Digital 7.31.FD1to .... [10503]
I'm sure others will have good suggestions but here's mine.
Increase LRU/CLEANERS significantly, like maybe 64 or more
noticed NETTYPE. Put your most used access method on CPU and the other on
NET
noticed NETTYPE For soctcp was 8,50. suggest trying 3,125
There are various onstat commands to examine which threads are the most
active. If you not using KAIO you may want to up AIO threads.
****************************************************************************
***
Forum Note: Use "Reply" to post a response in the discussion forum.
KAIO, yes i hear from that but i dont know how it works with this Compaq Alpha
ES40 machine on Compaq True64 unix???
My problem is my expirience with all this is more than 8 years ago and at
moment i have no specialist for this sensitive things and i not want make
experiments on the productivesystem!
I make onstat -g ioq and see that the gfds have all 0.
I am dark remind that this means that KAIO is turned on????
Greetings
> -----Original Message-----
> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
> ROBERT LANGHAMMER
> Sent: Wednesday, November 28, 2007 10:38 AM
> To: ids@iiug.org
> Subject: Re: RE: Update from Digital 7.31.FD1to .... [10508]
>
> KAIO, yes i hear from that but i dont know how it works with this
Compaq
> Alpha
> ES40 machine on Compaq True64 unix???
>
> My problem is my expirience with all this is more than 8 years ago and
at
> moment i have no specialist for this sensitive things and i not want
make
> experiments on the productivesystem!
>
> I make onstat -g ioq and see that the gfds have all 0.
> I am dark remind that this means that KAIO is turned on????
If KAIO is turned on, in the output from your onstat -g ioq command, you
should see a kio thread that looks like this:
IBM Informix Dynamic Server Version 9.40.FC8 -- Read-Only (Sec) --
Up 5 days 00:22:29 -- 598836 Kbytes
AIO I/O queues:q name/id len maxlen totalops dskread dskwrite dskcopy
sqli_dbg 0 0 0 0 0 0 0
kio 0 0 34 8647160 2025015 6622145 0
--EEM
>
> Greetings
>
>
>
************************************************************************
**
> *****
> Forum Note: Use "Reply" to post a response in the discussion forum.
It seems to me that it is turned on:
onstat -g ioq | more
Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 1 days 23:05:38 --
2039808 Kbytes
AIO I/O queues:
q name/id len maxlen totalops dskread dskwrite dskcopy
kio 0 0 32 14645648 13622478 1023170 0
kio 1 0 32 14571850 13686073 885777 0
kio 2 0 32 13797739 12946034 851705 0
adt 0 0 0 0 0 0 0
msc 0 0 2 82342 0 0 0
aio 0 0 2 63441 21123 0 0
pio 0 0 0 0 0 0 0
lio 0 0 0 0 0 0 0
gfd 3 0 0 0 0 0 0
.
.
.
.
Back to one of the origional issues, performace goes South when there are 150
users on the box. Believe this issue has been discussed before, how many
NETTYPE threads can be configured of each configured CPU. You have configured
3 CPU's but list:
NETTYPE soctcp,8,50,NET # Configure poll thread(s) for nettype
in the $ONCONFIG. Seems to me something like:
NETTYPE soctcp,3,80,CPU # Configure poll thread(s) for nettype
would give better performance. Anyone remember how this goes?
Baan IV run not on the same machine and connect to the informix server with an 1GB crossoverethernetcable. so we use not Shared Memory to access the database!
Robert
You have upgraded the server but not changed Informix?
SHMVIRTSIZE seems high, but I am unfamiliar with your platform and
application. Increasing BUFFERS would have been a better start point.
Do you have an issue with CPU, disk IO or memory on your server?
Does the application run on your server or only Informix?
Return
onstat -p
onstat -g seg
onstat -d
onstat -m
onstat -D
onstat -g iof
MW
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
ROBERT LANGHAMMER
Sent: Wednesday, 28 November 2007 9:12 p.m.
To: ids@iiug.org
Subject: Re: RE: Update from Digital 7.31.FD1to .... [10498]
Okay here the machine data and onconfig:
The machine is a Compaq True 64 Unix V4.0 1229 alpha with 4 CPUs and 4
GB
mainmemory.
The only changes what we make is what stand in the releasenotes:
shmmax = not change (2147483647)
shmmni from 512 to 2048
shmseg from 150 to 512
semmni 400 (not changed)
semmsl 1600 (not changed)
semopm 400 (not changed)
semume 400 (not changed)
maxdsiz ( max-per-proc-data-size ) should change to 9223372036854775807
but
kernel let max 4398046511104
maxvas (vm-maxvas) from 2048000000 to 9223372036854775807
per-proc-adress-space (per-proc-address-space) from 2147483647 to
4398046511104
max-per-proc-address-space (max-per-proc-address-space) from 2147483647
to
4398046511104
onconfig are the same than before (temporary we change SHMVIRTSIZE from
1100000 to 3317760 but the admin befor make to smal swap space and we
get
problems so i turn this parameter back. now i put one more disk inside
and
make now biger swap space. (more than double mainmemory)
onconfig:
#***********************************************************************
***
#
# INFORMIX SOFTWARE, INC.
#
# Title: onconfig.baan
# Description: INFORMIX-OnLine Configuration Parameters
#
#***********************************************************************
***
# Root Dbspace Configuration
ROOTNAME rootdbs # Root dbspace name
ROOTPATH /dev/syschunk # Path for device containing root dbspace
ROOTOFFSET 0 # Offset of root dbspace into device (Kbytes)
ROOTSIZE 40000 # Size of root dbspace (Kbytes)
# Disk Mirroring Configuration Parameters
MIRROR 0 # Mirroring flag (Yes = 1, No = 0)
MIRRORPATH # Path for device containing mirrored root
MIRROROFFSET 0 # Offset into mirrored device (Kbytes)
# Physical Log Configuration
PHYSDBS plogdbs # Location (dbspace) of physical log
PHYSFILE 299000 # Physical log file size (Kbytes)
# Logical Log Configuration
LOGFILES 100 # Number of logical log files
LOGSIZE 10000 # Logical log size (Kbytes)
# Diagnostics
MSGPATH /informix/online.log # System message log file path
CONSOLE /dev/console # System console message path
ALARMPROGRAM /informix/etc/log_full.sh # Alarm program path
#ALARMPROGRAM /informix/etc/no_log.sh # Alarm program path
TBLSPACE_STATS 1
# System Archive Tape Device
TAPEDEV /dev/null # Tape device path
TAPEBLK 64 # Tape block size (Kbytes)
TAPESIZE 61440000 # Maximum amount of data to put on tape (Kbytes)
# Log Archive Tape Device
LTAPEDEV /tmp/log.log # Log tape device path
#LTAPEDEV /dev/null # Log tape device path
LTAPEBLK 64 # Log tape block size (Kbytes)
LTAPESIZE 61440000 # Max amount of data to put on log tape (Kbytes)
# Optical
STAGEBLOB # INFORMIX-OnLine/Optical staging area
# System Configuration
SERVERNUM 1 # Unique id corresponding to a OnLine instance
DBSERVERNAME inf_baan # Name of default database server
DBSERVERALIASES inf_normal # List of alternate dbservernames
NETTYPE soctcp,8,50,NET # Configure poll thread(s) for nettype
NETTYPE ipcstr,1,60,CPU # Configure poll thread(s) for nettype
NETTYPE sqlmux # Configure poll thread(s) for nettype
DEADLOCK_TIMEOUT 60 # Max time to wait of lock in distributed env.
RESIDENT 0 # Forced residency flag (Yes = 1, No = 0)
MULTIPROCESSOR 1 # 0 for single-processor, 1 for multi-processor
NUMCPUVPS 3 # Number of user (cpu) vps
SINGLE_CPU_VP 0 # If non-zero, limit number of cpu vps to one
NOAGE 1 # Process aging
AFF_SPROC 1 # Affinity start processor
AFF_NPROCS 3 # Affinity number of processors
# Shared Memory Parameters
LOCKS 500000 # Maximum number of locks
BUFFERS 300000#BUFFERS 200000 # 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)LOGSMAX 100 # Maximum number of logical log files
CLEANERS 8 # Number of buffer cleaner processes
SHMBASE 0x200000000 # Shared memory base address
SHMVIRTSIZE 1100000#SHMVIRTSIZE 753664 # initial virtual shared memory segment size
SHMADD 57344 # Size of new shared memory segments (Kbytes)
SHMTOTAL 0 # Total shared memory (Kbytes). 0=>unlimited
CKPTINTVL 600 # Check point interval (in sec)
LRUS 8 # Number of LRU queues
LRU_MAX_DIRTY 5 # LRU percent dirty begin cleaning limit
LRU_MIN_DIRTY 2 # LRU percent dirty end cleaning limit
LTXHWM 50 # Long transaction high water mark percentage
LTXEHWM 60 # Long transaction high water mark (exclusive)
TXTIMEOUT 0x12c # Transaction timeout (in sec)
STACKSIZE 64 # Stack size (Kbytes)
# System Page Size
# BUFFSIZE - OnLine no longer supports this configuration parameter.
# To determine the page size used by OnLine on your platform
# see the last line of output from the command, 'onstat -b'.
# Recovery Variables
# OFF_RECVRY_THREADS:
# Number of parallel worker threads during fast recovery or an offline
restore.
# ON_RECVRY_THREADS:
# Number of parallel worker threads during an online restore.
OFF_RECVRY_THREADS 10 # Default number of offline worker threads
ON_RECVRY_THREADS 1 # Default number of online worker threads
# Data Replication Variables
# DRAUTO: 0 manual, 1 retain type, 2 reverse type
DRAUTO 0 # DR automatic switchover
DRINTERVAL 30 # DR max time between DR buffer flushes (in sec)
DRTIMEOUT 30 # DR network timeout (in sec)DRLOSTFOUND /informix/etc/dr.lostfound # DR lost+found file path
# CDR Variables
CDR_LOGBUFFERS 2048 # size of log reading buffer pool (Kbytes)
CDR_EVALTHREADS 1,2 # evaluator threads (per-cpu-vp,additional)
CDR_DSLOCKWAIT 5 # DS lockwait timeout (seconds)
CDR_QUEUEMEM 4096 # Maximum amount of memory for any CDR queue (Kbytes)
# Backup/Restore variables
BAR_BSALIB_PATH /usr/lib/libbsa.so
BAR_ACT_LOG /informix/bar_act.log
BAR_MAX_BACKUP 1
BAR_RETRY 1
BAR_NB_XPORT_COUNT 10
BAR_XFER_BUF_SIZE 31
ISM_DATA_POOL RocoDaten
ISM_LOG_POOL RocoLogs
# Read Ahead Variables
RA_PAGES 72 # Number of pages to attempt to read ahead
RA_THRESHOLD 66 # Number of pages left before next group
# DBSPACETEMP:
# OnLine equivalent of DBTEMP for SE. This is the list of dbspaces
# that the OnLine SQL Engine will use to create temp tables etc.
# If specified it must be a colon separated list of dbspaces that exist
# when the OnLine system is brought online. If not specified, or if
# all dbspaces specified are invalid, vario
>Baan IV run not on the same machine and connect to the informix server with an >1GB crossoverethernetcable. >so we use not Shared Memory to access the database! NETTYPE soctcp refers to tcpip connections which your Bann connections use Generally speaking, put the connection method most used on a "CPU" type VP. Your $ONCONFIG has soctcp running in a "NET" thread. It may be no accident that you started having issues when the user count reached 150. If you can only have one soctcp thread on each configured CPU then 3 * 50 = the max number of users that can be reliably serviced. I'm not sure informix will start multiple services threads on a CPU. Your $ONCONFIG calls for 8 threads but maybe it's only starting one for each configured CPU. Not long ago this issue of how user threads are serviced was discussed in this venue. I don't remember the outcome but your symptoms seems to be related to it. Given the number of users I'm sure other configuration changes could be made to inhance performace but any restriction as to how many users can attach or how efficent BUFFERS are managed (LRUs/CLEANERS) is the first place I would start.
RALPH GENTRY wrote: I'm going to disagree with Ralph. I hold strongly that TCP connections listener/poll threads should NEVER be configured to run in CPU VPs ONLY in NET VPs. I discussed it in depth in the Informix FAQ among other places, but the Q&D is in the CPU VP TCP poll threads have to burn CPU cycles to check the ports because the CPU VPs cannot be blocked to listen, they have work to do! And of course, while they are polling, they're not querying! You'll be using up CPU cycles that could be more usefully used to get work done. Art S. Kagel >> Baan IV run not on the same machine and connect to the informix server with >> > an > >> 1GB crossoverethernetcable. >> so we use not Shared Memory to access the database! >> > > NETTYPE soctcp refers to tcpip connections which your Bann connections use > > Generally speaking, put the connection method most used on a "CPU" type VP. > Your $ONCONFIG has soctcp running in a "NET" thread. It may be no accident > that you started having issues when the user count reached 150. If you can > only have one soctcp thread on each configured CPU then 3 * 50 = the max > number of users that can be reliably serviced. I'm not sure informix will > start multiple services threads on a CPU. Your $ONCONFIG calls for 8 threads > but maybe it's only starting one for each configured CPU. Not long ago this > issue of how user threads are serviced was discussed in this venue. I don't > remember the outcome but your symptoms seems to be related to it. > > Given the number of users I'm sure other configuration changes could be made > to inhance performace but any restriction as to how many users can attach or > how efficent BUFFERS are managed (LRUs/CLEANERS) is the first place I would > start. > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > >
onstat -p
Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 2 days 19:15:40 --
2039808 Kbytes
Profile
dskreads pagreads bufreads %cached dskwrits pagwrits bufwrits %cached
57290910 204865922 33221652373 99.83 4158407 37102592 18757603 77.83
isamtot open start read write rewrite delete commit rollbk
67619352616 32820763 268090029 30179662322 2303489 1009986 1091525 1093811 62
gp_read gp_write gp_rewrt gp_del gp_alloc gp_free gp_curs
0 0 0 0 0 0 0
ovlock ovuserthread ovbuff usercpu syscpu numckpts flushes
0 0 0 415904.13 12668.70 401 802
bufwaits lokwaits lockreqs deadlks dltouts ckpwaits compress seqscans
7929150 109 431756709 0 0 803 263556 741805
ixda-RA idx-RA da-RA RA-pgsused lchwaits
16862601 14500295 3733907 34174026 1920881
***********
>onstat -g seg
Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 2 days 19:16:38 --
2039808 Kbytes
Segment Summary:
id key addr size ovhd class blkused blkfree
1 1381451777 200000000 721420288 40480 R 87379 685
2 1381451778 22b000000 1132462080 17840 V 124683 13557
3 1381451779 26e800000 58720256 1456 V 6094 1074
4 1381451780 272000000 58720256 1456 V 6052 1116
5 1381451781 275800000 58720256 1456 V 5283 1885
6 1381451782 279000000 58720256 1456 V 1761 5407
Total: - - 2088763392 - - 231252 23724
********
onstat -d
Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 2 days 19:18:20 --
2039808 Kbytes
Dbspaces
address number flags fchunk nchunks flags owner name
22b07e1c0 1 1 1 1 N informix rootdbs
22bd34a00 2 1 2 2 N informix llogdbs
22bd34ae8 3 1 3 1 N informix plogdbs
22bd34bd0 4 2001 4 2 N T informix tempdbs
22bd34cb8 5 1 5 27 N informix datdbs
22bd34da0 6 2001 12 2 N T informix tempdbs1
22bd34e88 7 1 16 15 N informix testdbs
22bd34f70 8 1 27 3 N informix rechnungswesen
22bd35058 9 1 50 1 N informix archiv
9 active, 2047 maximum
Chunks
address chk/dbs offset size free bpages flags pathname
22b07e2a8 1 1 0 20000 4529 PO- /dev/syschunk
22b07e9c0 2 2 20000 200000 4947 PO- /dev/syschunk
22b07ead8 3 3 220000 150000 447 PO- /dev/syschunk
22b07ebf0 4 4 370000 200000 184405 PO- /dev/syschunk
22b07ed08 5 5 50 1000000 75 PO- /dev/datchunk18
22b07ee20 6 5 50 1000000 267 PO- /dev/datchunk25
22b07ef38 7 5 50 1000000 500 PO- /dev/datchunk10
22b07f050 8 5 50 1000000 0 PO- /dev/datchunk14
22b07f168 9 5 50 1000000 464 PO- /dev/datchunk31
22b07f280 10 5 50 1000000 104 PO- /dev/datchunk6
22b07f398 11 5 0 1000000 191 PO- /dev/datchunk19
22b07f4b0 12 6 0 313000 290629 PO- /dev/datchunk17
22b07f5c8 13 5 0 1000000 860 PO- /dev/datchunk26
22b07f6e0 14 5 0 1000000 322 PO- /dev/datchunk11
22b07f7f8 15 5 0 1000000 9324 PO- /dev/datchunk15
22b07f910 16 7 0 1000000 148985 PO- /dev/datchunk20
22b07fa28 17 7 0 1000000 433 PO- /dev/datchunk27
22b07fb40 18 7 0 1000000 151 PO- /dev/datchunk12
22b07fc58 19 7 0 1000000 4 PO- /dev/datchunk16
22b07fd70 20 7 0 1000000 1 PO- /dev/datchunk32
22b07fe88 21 7 0 1000000 580797 PO- /dev/datchunk7
22b0b8850 22 5 0 1000000 11148 PO- /dev/datchunk8
22b0b8968 23 5 0 1000000 12445 PO- /dev/datchunk21
22b0b8a80 24 5 0 1000000 12551 PO- /dev/datchunk28
22b0b8b98 25 5 0 1000000 890 PO- /dev/datchunk22
22b0b8cb0 26 7 0 1000000 263851 PO- /dev/datchunk23
22b0b8dc8 27 8 0 1000000 789 PO- /dev/datchunk2
22b0b8ee0 28 5 0 1000000 5669 PO- /dev/datchunk4
22b0b8ff8 29 7 0 1000000 16934 PO- /dev/datchunk24
22b0b9110 30 5 0 5000 121 PO- /dev/datchunk33
22b0b9228 31 5 0 1000000 12244 PO- /dev/datchunk29
22b0b9340 32 5 0 1000000 11573 PO- /dev/datchunk30
22b0b9458 33 5 25 1000000 12559 PO- /dev/datchunk34
22b0b9570 34 5 25 1000000 8278 PO- /dev/datchunk41
22b0b9688 35 5 25 1000000 5095 PO- /dev/datchunk48
22b0b97a0 36 5 0 1000000 27898 PO- /dev/datchunk55
22b0b98b8 37 8 0 1000000 30925 PO- /dev/datchunk36
22b0b99d0 38 6 5000 235000 234997 PO- /dev/datchunk33
22b0b9ae8 39 4 240000 231000 230997 PO- /dev/datchunk33
22b0b9c00 40 5 0 1000000 14503 PO- /dev/datchunk62
22b0b9d18 41 5 0 1000000 133897 PO- /dev/datchunk42
22b0b9e30 42 8 0 1000000 999997 PO- /dev/datchunk35
22b0c9c50 43 7 0 1000000 18247 PO- /dev/datchunk5
22b0c9d68 44 7 0 1000000 5953 PO- /dev/datchunk37
22b0c9e80 45 7 0 1000000 35599 PO- /dev/datchunk43
22bd34028 46 7 0 1000000 340600 PO- /dev/datchunk49
22bd34140 47 7 0 1000000 550323 PO- /dev/datchunk56
22bd34258 48 7 0 1000000 985260 PO- /dev/datchunk44
22bd34370 49 7 0 1000000 999997 PO- /dev/datchunk50
22bd34488 50 9 0 1000000 223157 PO- /dev/datchunk38
22bd345a0 51 5 0 1000000 60421 PO- /dev/datchunk57
22bd346b8 52 5 0 1000000 180439 PO- /dev/datchunk63
22bd347d0 53 5 0 1000000 14533 PO- /dev/datchunk58
22bd348e8 54 2 0 1000000 694997 PO- /dev/datchunk45
54 active, 2047 maximum
************
onstat -m
Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 2 days 19:18:51 --
2039808 Kbytes
Message Log File: /informix/online.log
13:47:04 Checkpoint loguniq 360397, logpos 0x36b018
13:57:08 Checkpoint Completed: duration was 5 seconds.
13:57:08 Checkpoint loguniq 360397, logpos 0x11d2018
14:00:21 Logical Log 360397 Complete.
14:00:22 Logical Log 360397 - Backup Started
14:00:39 Logical Log 360397 - Backup Completed
14:07:11 Checkpoint Completed: duration was 1 seconds.
14:07:11 Checkpoint loguniq 360398, logpos 0xa0f14c
14:13:46 Logical Log 360398 Complete.
14:13:47 Logical Log 360398 - Backup Started
14:14:08 Logical Log 360398 - Backup Completed
14:17:12 Checkpoint Completed: duration was 1 seconds.
14:17:12 Checkpoint loguniq 360399, logpos 0x240018
14:27:14 Checkpoint Completed: duration was 1 seconds.
14:27:14 Checkpoint loguniq 360399, logpos 0xd8c018
*************
onstat -D
Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 2 days 19:19:11 --
2039808 Kbytes
Dbspaces
address number flags fchunk nchunks flags owner name
22b07e1c0 1 1 1 1 N informix rootdbs
22bd34a00 2 1 2 2 N informix llogdbs
22bd34ae8 3 1 3 1 N informix plogdbs
22bd34bd0 4 2001 4 2 N T informix tempdbs
22bd34cb8 5 1 5 27 N informix datdbs
22bd34da0 6 2001 12 2 N T informix tempdbs1
22bd34e88 7 1 16 15 N informix testdbs
22bd34f70 8 1 27 3 N informix rechnungswesen
22bd35058 9 1 50 1 N informix archiv
9 active, 2047 maximum
Chunks
address chk/dbs offset page Rd page Wr pathname
22b07e2a8 1 1 0 80788 5018 /dev/syschunk
22b07e9c0 2 2 20000 544471 542705 /dev/syschunk
22b07ead8 3 3 220000 27 1129461 /dev/syschunk
22b07ebf0 4 4 370000 14665828 16276107 /dev/syschunk
22b07ed08 5 5 50 3330194 63954 /dev/datchunk18
22b07ee20 6 5 50 3693226 91485 /dev/datchunk25
22b07ef38 7 5 50 3924806 54026 /dev/datchunk10
22b07f050 8 5 50 3005819 5146 /dev/datchunk14
22b07f168 9 5 50 2556638 60355 /dev/datchunk31
22b07f280 10 5 50 6149285 64578 /dev/datchunk6
22b07f398 11 5 0 3351048 81633 /dev/datchunk19
22b07f4b0 12 6 0 14660318 16530324 /dev/datchunk17
22b07f5c8 13 5 0 5369868 116983 /dev/datchunk26
22b07f6e0 14 5 0 2756039 18098 /dev/datchunk11
22b07f7f8 15 5 0 3751442 100414 /dev/datchunk15
22b07f910 16 7 0 2519886 0 /dev/datchunk20
22b07fa28 17 7 0 3020597 0 /dev/datchunk27
22b07fb40 18 7 0 2961168 0 /dev/datchunk12
22b07fc58 19 7
ROBERT LANGHAMMER wrote:
I think that your biggest problem is your disk layout. You have rootdb,
physical logs, logical logs and one of your temp dbspaces all defined on
the same device /dev/syschunk and you are POUNDING on the tempdb
spaces. A few other chunks are being hit hard also, namely
/dev/datachunk6, 17, 26, 21,22,4,29,30,34,41,48,55,62, & 63. Those 18
chunks (including the four chunks defined on syschunk) are each being
hit with over 4 IO/sec (note that IDS averages IOs over clock time not
elapsed time so these averages are VERY low if they could be compared to
peak IO rates). I hope they are not all defined on the same physical
structure (ie stripe, RAID array, NAS, SAN, etc.) since these rates
indicate that at peak you would likely swamp the throughput capacity of
a single structure and it's channel.
I also note the VERY high temp dbspace activity as highly unusual. Is
is possible that your data distributions are not sufficient or that
there are missing indexes? Are your queries relying heavily on VIEWs,
especially joining VIEWs to other VIEWS and TABLES? Such operations
create temp tables containing intermediate results - notably the results
of the VIEW query to be joined to other results. Do you have ORDER BY
clauses not supported by indexes? If the applications are interactive,
should you be running the queries using FIRST_ROWS optimization to avoid
sorts? Can you increase DBUPSPACE or enable PDQPRIORITY and parallel
sorting (PSORT_NPROCS & PSORT_DBTEMP) to enable in-memory sorting
reducing disk IO loads and improve responsiveness?
Art S. Kagel
> <SNIP>
> onstat -D>
> Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 2 days 19:19:11 --
> 2039808 Kbytes
>
> Dbspaces
> address number flags fchunk nchunks flags owner name
> 22b07e1c0 1 1 1 1 N informix rootdbs
> 22bd34a00 2 1 2 2 N informix llogdbs
> 22bd34ae8 3 1 3 1 N informix plogdbs
> 22bd34bd0 4 2001 4 2 N T informix tempdbs
> 22bd34cb8 5 1 5 27 N informix datdbs
> 22bd34da0 6 2001 12 2 N T informix tempdbs1
> 22bd34e88 7 1 16 15 N informix testdbs
> 22bd34f70 8 1 27 3 N informix rechnungswesen
> 22bd35058 9 1 50 1 N informix archiv
> 9 active, 2047 maximum
>
> Chunks
> address chk/dbs offset page Rd page Wr pathname
> 22b07e2a8 1 1 0 80788 5018 /dev/syschunk
> 22b07e9c0 2 2 20000 544471 542705 /dev/syschunk
> 22b07ead8 3 3 220000 27 1129461 /dev/syschunk
> 22b07ebf0 4 4 370000 14665828 16276107 /dev/syschunk
> 22b07ed08 5 5 50 3330194 63954 /dev/datchunk18
> 22b07ee20 6 5 50 3693226 91485 /dev/datchunk25
> 22b07ef38 7 5 50 3924806 54026 /dev/datchunk10
> 22b07f050 8 5 50 3005819 5146 /dev/datchunk14
> 22b07f168 9 5 50 2556638 60355 /dev/datchunk31
> 22b07f280 10 5 50 6149285 64578 /dev/datchunk6
> 22b07f398 11 5 0 3351048 81633 /dev/datchunk19
> 22b07f4b0 12 6 0 14660318 16530324 /dev/datchunk17
> 22b07f5c8 13 5 0 5369868 116983 /dev/datchunk26
> 22b07f6e0 14 5 0 2756039 18098 /dev/datchunk11
> 22b07f7f8 15 5 0 3751442 100414 /dev/datchunk15
> 22b07f910 16 7 0 2519886 0 /dev/datchunk20
> 22b07fa28 17 7 0 3020597 0 /dev/datchunk27
> 22b07fb40 18 7 0 2961168 0 /dev/datchunk12
> 22b07fc58 19 7 0 3006185 0 /dev/datchunk16
> 22b07fd70 20 7 0 3002416 0 /dev/datchunk32
> 22b07fe88 21 7 0 1199256 0 /dev/datchunk7
> 22b0b8850 22 5 0 4051669 51109 /dev/datchunk8
> 22b0b8968 23 5 0 4652059 37607 /dev/datchunk21
> 22b0b8a80 24 5 0 4286159 37021 /dev/datchunk28
> 22b0b8b98 25 5 0 6271816 25651 /dev/datchunk22
> 22b0b8cb0 26 7 0 2185444 0 /dev/datchunk23
> 22b0b8dc8 27 8 0 2887002 9856 /dev/datchunk2
> 22b0b8ee0 28 5 0 4735443 49371 /dev/datchunk4
> 22b0b8ff8 29 7 0 2904982 0 /dev/datchunk24
> 22b0b9110 30 5 0 23896 23 /dev/datchunk33
> 22b0b9228 31 5 0 6504509 46304 /dev/datchunk29
> 22b0b9340 32 5 0 7621371 44343 /dev/datchunk30
> 22b0b9458 33 5 25 8388770 84764 /dev/datchunk34
> 22b0b9570 34 5 25 6839096 47487 /dev/datchunk41
> 22b0b9688 35 5 25 7417666 37712 /dev/datchunk48
> 22b0b97a0 36 5 0 11433054 38697 /dev/datchunk55
> 22b0b98b8 37 8 0 3440784 32119 /dev/datchunk36
> 22b0b99d0 38 6 5000 43684 104921 /dev/datchunk33
> 22b0b9ae8 39 4 240000 1 1 /dev/datchunk33
> 22b0b9c00 40 5 0 9674343 352861 /dev/datchunk62
> 22b0b9d18 41 5 0 1600315 27613 /dev/datchunk42
> 22b0b9e30 42 8 0 7 0 /dev/datchunk35
> 22b0c9c50 43 7 0 2927830 0 /dev/datchunk5
> 22b0c9d68 44 7 0 2797774 0 /dev/datchunk37
> 22b0c9e80 45 7 0 2746234 0 /dev/datchunk43
> 22bd34028 46 7 0 1930641 0 /dev/datchunk49
> 22bd34140 47 7 0 1211398 0 /dev/datchunk56
> 22bd34258 48 7 0 42601 0 /dev/datchunk44
> 22bd34370 49 7 0 7 0 /dev/datchunk50
> 22bd34488 50 9 0 1951771 0 /dev/datchunk38
> 22bd345a0 51 5 0 3151076 65447 /dev/datchunk57
> 22bd346b8 52 5 0 6366288 36740 /dev/datchunk63
> 22bd347d0 53 5 0 2517689 60185 /dev/datchunk58
> 22bd348e8 54 2 0 867594 871397 /dev/datchunk45
> 54 active, 2047 maximum
> ************
> onstat -g iof>
> Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 2 days 19:19:46 --
> 2039808 Kbytes
>
> AIO global files:
> gfd pathname totalops dskread dskwrite io/s
> 3 /dev/syschunk 66682 18 66664 0.3
> 4 /dev/syschunk 482673 17676 464997 2.0
> 5 /dev/syschunk 1162575 570984 591591 4.8
> 6 /dev/syschunk 29021 25161 3860 0.1
> 7 /dev/datchunk18 520232 472544 47688 2.1
> 8 /dev/datchunk25 673667 605816 67851 2.8
> 9 /dev/datchunk10 654100 611712 42388 2.7
> 10 /dev/datchunk14 99915 99328 587 0.4
> 11 /dev/datchunk31 508035 486155 21880 2.1
> 12 /dev/datchunk6 1952558 1898436 54122 8.1
> 13 /dev/datchunk19 377096 333247 43849 1.6
> 14 /dev/datchunk17 1174340 572236 602104 4.8
> 15 /dev/datchunk26 1668881 1588663 80218 6.9
> 16 /dev/datchunk11 208547 195066 13481 0.9
> 17 /dev/datchunk15 804755 728195 76560 3.3
> 18 /dev/datchunk20 125034 125034 0 0.5
> 19 /dev/datchunk27 117503 117503 0 0.5
> 20 /dev/datchunk12 94194 94194 0 0.4
> 21 /dev/datchunk16 101402 101402 0 0.4
> 22 /dev/datchunk32 96175 96175 0 0.4
> 23 /dev/datchunk7 52770 52770 0 0.2
> 24 /dev/datchunk8 938449 899371 39078 3.9
> 25 /dev/datchunk21 1092558 1061818 30740 4.5
> 26 /dev/datchunk28 848405 816846 31559 3.5
> 27 /dev/datchunk22 1258845 1244243 14602 5.2
> 28 /dev/datchunk23 69439 69439 0 0.3
> 29 /dev/datchunk2 240657 235815 4842 1.0
> 30 /dev/datchunk4 1656283 1619342 36941 6.8
> 31 /dev/datchunk24 91429 91429 0 0.4
> 32 /dev/datchunk33 7833 7812 21 0.0
> 33 /dev/datchunk29 2009699 1974065 35634 8.3
> 34 /dev/datchunk30 2882732 2848402 34330 11.9
> 35 /dev/datchunk34 3222037 3160275 61762 13.3
> 36 /dev/datchunk41 2400999 2367021 33978 9.9
> 37 /dev/datchunk48 2963750 2938241 25509 12.2
> 38 /dev/datchunk55 5575715 5543933 31782 23.0
> 39 /dev/datchunk36 394250 369734 24516 1.6
> 40 /dev/datchunk33 5316 1564 3752 0.0
> 41 /dev/datchunk33 2 1 1 0.0
> 42 /dev/datchunk62 2837455 2642187 195268 11.7
> 43 /dev/datchunk42 69913 63961 5952 0.3
> 44 /dev/datchunk35 7 7 0 0.0
> 45 /dev/datchunk5 92389 92389 0 0.4
> 46 /dev/datchunk37 96097 96097 0 0.4
> 47 /dev/datchunk43 86065 86065 0 0.4
> 48 /dev/datchunk49 60477 60477 0 0.2
> 49 /dev/datchunk56 38230 38230 0 0.2
> 50 /dev/datchunk44 1360 1360 0 0.0
> 51 /dev/datchunk50 7 7 0 0.0
> 5
In addition to Arts comments.
If my calculations are correct you have 3CPUVPS and your usercpu and
syscpu add up to the equivalent of almost 2 processes running full time
over 2 days +. What is your CPU use and disk IO when the system is
busy?
There does not appear to be a large number of sequential scans, but you
may need to look for these in long running SQLS..
onstat -u
onstat -g ses {number}
You have 5 type V shared memory segments. Increase SHMVIRTSIZE so that
additional segments are not created. What is running when more segments
are added? There may be some issue with that application.
You have Informix running in under 2Gb on a server with 4Gb. Adding
more BUFFERS could help your IO but hinder checkpoints (add another
500000 to allocate another 1Gb). You may want to increase CKPTINTVL to
1200 or decrease LRU_MAX_DIRTY to 2 and LRU_MIN_DIRTY to 1.
MW
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
ROBERT LANGHAMMER
Sent: Friday, 30 November 2007 2:31 a.m.
To: ids@iiug.org
Subject: Re: RE: RE: Update from Digital 7.31.FD1to .... [10530]
onstat -p
Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 2 days
19:15:40 --
2039808 Kbytes
Profile
dskreads pagreads bufreads %cached dskwrits pagwrits bufwrits %cached
57290910 204865922 33221652373 99.83 4158407 37102592 18757603 77.83
isamtot open start read write rewrite delete commit rollbk
67619352616 32820763 268090029 30179662322 2303489 1009986 1091525
1093811 62
gp_read gp_write gp_rewrt gp_del gp_alloc gp_free gp_curs
0 0 0 0 0 0 0
ovlock ovuserthread ovbuff usercpu syscpu numckpts flushes
0 0 0 415904.13 12668.70 401 802
bufwaits lokwaits lockreqs deadlks dltouts ckpwaits compress seqscans
7929150 109 431756709 0 0 803 263556 741805
ixda-RA idx-RA da-RA RA-pgsused lchwaits
16862601 14500295 3733907 34174026 1920881
***********
>onstat -g seg
Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 2 days
19:16:38 --
2039808 Kbytes
Segment Summary:
id key addr size ovhd class blkused blkfree
1 1381451777 200000000 721420288 40480 R 87379 685
2 1381451778 22b000000 1132462080 17840 V 124683 13557
3 1381451779 26e800000 58720256 1456 V 6094 1074
4 1381451780 272000000 58720256 1456 V 6052 1116
5 1381451781 275800000 58720256 1456 V 5283 1885
6 1381451782 279000000 58720256 1456 V 1761 5407
Total: - - 2088763392 - - 231252 23724
********
onstat -d
Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 2 days
19:18:20 --
2039808 Kbytes
Dbspaces
address number flags fchunk nchunks flags owner name
22b07e1c0 1 1 1 1 N informix rootdbs
22bd34a00 2 1 2 2 N informix llogdbs
22bd34ae8 3 1 3 1 N informix plogdbs
22bd34bd0 4 2001 4 2 N T informix tempdbs
22bd34cb8 5 1 5 27 N informix datdbs
22bd34da0 6 2001 12 2 N T informix tempdbs1
22bd34e88 7 1 16 15 N informix testdbs
22bd34f70 8 1 27 3 N informix rechnungswesen
22bd35058 9 1 50 1 N informix archiv
9 active, 2047 maximum
Chunks
address chk/dbs offset size free bpages flags pathname
22b07e2a8 1 1 0 20000 4529 PO- /dev/syschunk
22b07e9c0 2 2 20000 200000 4947 PO- /dev/syschunk
22b07ead8 3 3 220000 150000 447 PO- /dev/syschunk
22b07ebf0 4 4 370000 200000 184405 PO- /dev/syschunk
22b07ed08 5 5 50 1000000 75 PO- /dev/datchunk18
22b07ee20 6 5 50 1000000 267 PO- /dev/datchunk25
22b07ef38 7 5 50 1000000 500 PO- /dev/datchunk10
22b07f050 8 5 50 1000000 0 PO- /dev/datchunk14
22b07f168 9 5 50 1000000 464 PO- /dev/datchunk31
22b07f280 10 5 50 1000000 104 PO- /dev/datchunk6
22b07f398 11 5 0 1000000 191 PO- /dev/datchunk19
22b07f4b0 12 6 0 313000 290629 PO- /dev/datchunk17
22b07f5c8 13 5 0 1000000 860 PO- /dev/datchunk26
22b07f6e0 14 5 0 1000000 322 PO- /dev/datchunk11
22b07f7f8 15 5 0 1000000 9324 PO- /dev/datchunk15
22b07f910 16 7 0 1000000 148985 PO- /dev/datchunk20
22b07fa28 17 7 0 1000000 433 PO- /dev/datchunk27
22b07fb40 18 7 0 1000000 151 PO- /dev/datchunk12
22b07fc58 19 7 0 1000000 4 PO- /dev/datchunk16
22b07fd70 20 7 0 1000000 1 PO- /dev/datchunk32
22b07fe88 21 7 0 1000000 580797 PO- /dev/datchunk7
22b0b8850 22 5 0 1000000 11148 PO- /dev/datchunk8
22b0b8968 23 5 0 1000000 12445 PO- /dev/datchunk21
22b0b8a80 24 5 0 1000000 12551 PO- /dev/datchunk28
22b0b8b98 25 5 0 1000000 890 PO- /dev/datchunk22
22b0b8cb0 26 7 0 1000000 263851 PO- /dev/datchunk23
22b0b8dc8 27 8 0 1000000 789 PO- /dev/datchunk2
22b0b8ee0 28 5 0 1000000 5669 PO- /dev/datchunk4
22b0b8ff8 29 7 0 1000000 16934 PO- /dev/datchunk24
22b0b9110 30 5 0 5000 121 PO- /dev/datchunk33
22b0b9228 31 5 0 1000000 12244 PO- /dev/datchunk29
22b0b9340 32 5 0 1000000 11573 PO- /dev/datchunk30
22b0b9458 33 5 25 1000000 12559 PO- /dev/datchunk34
22b0b9570 34 5 25 1000000 8278 PO- /dev/datchunk41
22b0b9688 35 5 25 1000000 5095 PO- /dev/datchunk48
22b0b97a0 36 5 0 1000000 27898 PO- /dev/datchunk55
22b0b98b8 37 8 0 1000000 30925 PO- /dev/datchunk36
22b0b99d0 38 6 5000 235000 234997 PO- /dev/datchunk33
22b0b9ae8 39 4 240000 231000 230997 PO- /dev/datchunk33
22b0b9c00 40 5 0 1000000 14503 PO- /dev/datchunk62
22b0b9d18 41 5 0 1000000 133897 PO- /dev/datchunk42
22b0b9e30 42 8 0 1000000 999997 PO- /dev/datchunk35
22b0c9c50 43 7 0 1000000 18247 PO- /dev/datchunk5
22b0c9d68 44 7 0 1000000 5953 PO- /dev/datchunk37
22b0c9e80 45 7 0 1000000 35599 PO- /dev/datchunk43
22bd34028 46 7 0 1000000 340600 PO- /dev/datchunk49
22bd34140 47 7 0 1000000 550323 PO- /dev/datchunk56
22bd34258 48 7 0 1000000 985260 PO- /dev/datchunk44
22bd34370 49 7 0 1000000 999997 PO- /dev/datchunk50
22bd34488 50 9 0 1000000 223157 PO- /dev/datchunk38
22bd345a0 51 5 0 1000000 60421 PO- /dev/datchunk57
22bd346b8 52 5 0 1000000 180439 PO- /dev/datchunk63
22bd347d0 53 5 0 1000000 14533 PO- /dev/datchunk58
22bd348e8 54 2 0 1000000 694997 PO- /dev/datchunk45
54 active, 2047 maximum
************
onstat -m
Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 2 days
19:18:51 --
2039808 Kbytes
Message Log File: /informix/online.log
13:47:04 Checkpoint loguniq 360397, logpos 0x36b018
13:57:08 Checkpoint Completed: duration was 5 seconds.
13:57:08 Checkpoint loguniq 360397, logpos 0x11d2018
14:00:21 Logical Log 360397 Complete.
14:00:22 Logical Log 360397 - Backup Started
14:00:39 Logical Log 360397 - Backup Completed
14:07:11 Checkpoint Completed: duration was 1 seconds.
14:07:11 Checkpoint loguniq 360398, logpos 0xa0f14c
14:13:46 Logical Log 360398 Complete.
14:13:47 Logical Log 360398 - Backup Started
14:14:08 Logical Log 360398 - Backup Completed
14:17:12 Checkpoint Completed: duration was 1 seconds.
14:17:12 Checkpoint loguniq 360399, logpos 0x240018
14:27:14 Checkpoint Completed: duration was 1 seconds.
14:27:14 Checkpoint loguniq 360399, logpos 0xd8c018
*************
onstat -D
Informix Dynamic Server Version 7.31.FD7 -- On-Line -- Up 2 days
19:19:11 --
2039808 Kbytes
Dbspaces
address number flags fchunk nchunks flags owner name
22b07e1c0 1 1 1 1 N informix rootdbs
22bd34a00 2 1 2 2 N informix llogdbs
22bd34ae8 3 1 3 1 N informix plogdbs
22bd34bd0