- Community Home
- >
- Servers and Operating Systems
- >
- Operating Systems
- >
- Operating System - HP-UX
- >
- SuperDome x Informix
Categories
Company
Local Language
Forums
Discussions
Forums
- Data Protection and Retention
- Entry Storage Systems
- Legacy
- Midrange and Enterprise Storage
- Storage Networking
- HPE Nimble Storage
Discussions
Discussions
Discussions
Forums
Forums
Discussions
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
- BladeSystem Infrastructure and Application Solutions
- Appliance Servers
- Alpha Servers
- BackOffice Products
- Internet Products
- HPE 9000 and HPE e3000 Servers
- Networking
- Netservers
- Secure OS Software for Linux
- Server Management (Insight Manager 7)
- Windows Server 2003
- Operating System - Tru64 Unix
- ProLiant Deployment and Provisioning
- Linux-Based Community / Regional
- Microsoft System Center Integration
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Discussion Boards
Community
Resources
Forums
Blogs
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-26-2006 12:43 AM
тАО06-26-2006 12:43 AM
SuperDome x Informix
u125:/logs/conf# model
9000/800/SD32A
u125:/logs/conf# uname -a
HP-UX u125 B.11.23 U 9000/800 441251589 unlimited-user license
We try use HP-UX 11.11 or HP-UX 11.23.
We are trying migrate Informix from a server AIX to a server HP-UX, after migration the user complain about system
perfomance.
Then, some output commands were captured and migration was canceled.
Now, we developed an stress java program wich simulates the problem about 100 users accessing table with 210000 lines. We
have about 4000 users actually using the production system in a AIX box.
we have IDS running in the PA-RISC processor.
u122 (u122) /home/informix$onstat -
IBM Informix Dynamic Server Version 7.31.FD9PORT -- On-Line -- Up 19:46:19 -- 10922728 Kbytes
We use the same IDS configuration (NUMCPUVPS, BUFFERS, KAIO, ....) on both
servers (RP and Superdome) including kernel parameters.
We installed IDS 7.31 FD9 in the RP-4440 server and problem doesnt occurred. Then I reinstall the SuperDome Server like
RP-4440 server.
The problem doesn┬┤t occur with RP-4440 using SAN Storage ( HDS or Shark).
Informix Support Team (IBM) and HP are working on the problem, the 7.31.FD9PORT developed by Informix solve the problem only
if we use chunks in the internal disks. If we use chunks in the storage, the problem happen again.
We have two SuperDome Servers each one are connect for differents Storages Box (IBM Shark / Hitachi HDS).
Informix Support Team (IBM) and HP continue working together to solve the problem.
The same problem occurs with IDS 9.40. This is the reason we are looking for people who has IDS running into SuperDome Server
to know it was possible to solve the problem. This problem only occurs in the SuperDome Machine.
See output commands below, wich "S-" on second column represents mutex contention.
u122 (u122) /home/informix$onstat -u | grep S
IBM Informix Dynamic Server Version 7.31.FD9PORT -- On-Line -- Up 19:52:40 -- 10922728 Kbytes
c00000011ae1f268 S--PR-- 34 dsvcona - c00000011c3a8c28 0 1 829 1232
c00000011ae20048 S--PR-- 35 dsvcona - c00000011c3a8c28 0 1 462 1232
c00000011ae20738 S--PR-- 36 dsvcona - c00000011c3a8c28 0 1 180 1120
c00000011ae21518 S--PR-- 38 dsvcona - c00000011c3a8c28 0 1 21 1176
c00000011ae21c08 S--PR-- 39 dsvcona - c00000011c3a8c28 0 1 21 1176
c00000011ae222f8 S--PR-- 40 dsvcona - c00000011c3a8c28 0 1 19 1064
c00000011ae229e8 S--PR-- 41 dsvcona - c00000011c3a8c28 0 1 15 840
c00000011ae230d8 S--PR-- 42 dsvcona - c00000011c3a8c28 0 1 8 448
c00000011ae245a8 S--PR-- 45 dsvcona - c00000011c3a8c28 0 1 5 280
c00000011ae26168 S--PR-- 49 dsvcona - c00000011c3a8c28 0 1 2 112
c00000011ae26f48 S--PR-- 51 dsvcona - c00000011c3a8c28 0 1 2 112
c00000011ae27638 S--PR-- 52 dsvcona - c00000011c3a8c28 0 1 2 112
c00000011ae27d28 S--PR-- 53 dsvcona - c00000011c3a8c28 0 1 2 112
c00000011ae28418 S--PR-- 54 dsvcona - c00000011c3a8c28 0 1 1 56
c00000011ae28b08 S--PR-- 55 dsvcona - c00000011c3a8c28 0 1 1 56
c00000011ae291f8 S--PR-- 56 dsvcona - c00000011c3a8c28 0 1 1 56
c00000011ae298e8 S--PR-- 57 dsvcona - c00000011c3a8c28 0 1 1 56
c00000011ae29fd8 S--PR-- 58 dsvcona - c00000011c3a8c28 0 1 1 56
c00000011ae2a6c8 S--PR-- 59 dsvcona - c00000011c3a8c28 0 1 1 56
c00000011ae2bb98 S--PR-- 62 dsvcona - c00000011c3a8c28 0 1 1 56
c00000011ae2c978 S--PR-- 64 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae2d068 S--PR-- 65 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae2de48 S--PR-- 67 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae2e538 S--PR-- 68 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae2ec28 S--PR-- 69 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae2f318 S--PR-- 70 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae2fa08 S--PR-- 71 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae300f8 S--PR-- 72 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae307e8 S--PR-- 73 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae31cb8 S--PR-- 76 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae32a98 S--PR-- 78 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae33878 S--PR-- 80 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae34658 S--PR-- 82 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae34d48 S--PR-- 83 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae35438 S--PR-- 84 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae35b28 S--PR-- 85 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae36218 S--PR-- 86 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae36ff8 S--PR-- 88 dsvcona - c00000011c3a8c28 0 1 0 0
c00000011ae376e8 S--PR-- 89 dsvcona - c00000011c3a8c28 0 1 0 0
u122 (u122) /home/informix$onstat -g lmx
IBM Informix Dynamic Server Version 7.31.FD9PORT -- On-Line -- Up 19:52:45 -- 10922728 Kbytes
Locked mutexes:
mid addr name holder lkcnt waiter waittime
u122 (u122) /home/informix$onstat -g lmx
IBM Informix Dynamic Server Version 7.31.FD9PORT -- On-Line -- Up 19:52:47 -- 10922728 Kbytes
Locked mutexes:
mid addr name holder lkcnt waiter waittime
u122 (u122) /home/informix$onstat -g lmx
IBM Informix Dynamic Server Version 7.31.FD9PORT -- On-Line -- Up 19:52:48 -- 10922728 Kbytes
Locked mutexes:
mid addr name holder lkcnt waiter waittime
2054 c00000011c3a8c28 pt_c0002a 205 0 105 21
u122 (u122) /home/informix$onstat -g lmx
IBM Informix Dynamic Server Version 7.31.FD9PORT -- On-Line -- Up 19:52:49 -- 10922728 Kbytes
Locked mutexes:
mid addr name holder lkcnt waiter waittime
2054 c00000011c3a8c28 pt_c0002a 140 0 153 11
150 12
128 15
107 21
159 11
125 15
125 15
181 8
176 9
160 11
135 14
144 12
110 21
103 22
171 9
130 14
186 8
166 10
154 11
161 10
128 15
Anybody been down this road before?
Tks
Schimidt
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-26-2006 09:41 PM
тАО06-26-2006 09:41 PM
Re: SuperDome x Informix
I run 8 instances on 9.40.FC3 on a 2 npar SD32 (16 in each npar), plus xp12000, using raw, sym-linked logical volumes, hp-ux 11.11, 64 bit. No problems at all, but I never had more than 600 sessions in each instance on that box.
I have never before heard of anyone using 7.31 on a SD.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-26-2006 09:51 PM
тАО06-26-2006 09:51 PM
Re: SuperDome x Informix
Another idea I have is this: how big are your shared memory segments and how many do you have? Keep this number low, see this, my favourite thread:
http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=147349
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-26-2006 10:33 PM
тАО06-26-2006 10:33 PM
Re: SuperDome x Informix
I├В┬┤m attach file with kernel parameter.
We monitor systems calls from process oninit who create segment shared memory .
Here, if we are using internal disks, the problem doesn├В┬┤t occur:
/var/tmp/hp/without_hang/t.out
u122:/var/tmp/hp/without_hang# more t.out
( Attached to process 10000 ("oninit -v") [64-bit] )
1150977768.245431 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) [sleeping]
1150977768.834259 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) = 0
1150977769.844237 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) = 0
1150977770.334164 [10000]{55517} #1 Received signal 14, SIGALRM, in semop(), [caught], no siginfo
1150977770.334353 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) ERR#4 EINTR
1150977770.334592 [10000]{55517} #1 kill(10000, SIG#0) ... = 0
1150977770.334775 [10000]{55517} #1 times(0xc00000011c603b68) = 5908523
1150977770.334951 [10000]{55517} #1 alarm(3) ............. = 0
1150977770.854234 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) = 0
1150977771.864208 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) = 0
1150977772.874210 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) = 0
1150977773.344126 [10000]{55517} #1 Received signal 14, SIGALRM, in semop(), [caught], no siginfo
1150977773.344257 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) ERR#4 EINTR
1150977773.344438 [10000]{55517} #1 kill(10000, SIG#0) ... = 0
1150977773.344610 [10000]{55517} #1 times(0xc00000011c603b68) = 5908824
1150977773.344781 [10000]{55517} #1 alarm(3) ............. = 0
1150977773.884200 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) = 0
1150977774.894186 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) = 0
1150977775.904178 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) = 0
1150977776.354098 [10000]{55517} #1 Received signal 14, SIGALRM, in semop(), [caught], no siginfo
1150977776.354232 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) ERR#4 EINTR
1150977776.354409 [10000]{55517} #1 kill(10000, SIG#0) ... = 0
Here, if we are using external disks, when the problem occurs:
u122:/var/tmp/hp/with_hang# more t.out
( Attached to process 10000 ("oninit -v") [64-bit] )
1150978315.441252 [10000]{55517} #1 semop(16415, 0xc00000011efbe530, 2) [running]
1150978315.441783 [10000]{55517} #1 semop(16415, 0xc00000011efbe530, 2) = 0
1150978315.442272 [10000]{55517} #1 semop(16415, 0xc00000011efbe530, 2) = 0
1150978315.442734 [10000]{55517} #1 semop(16415, 0xc00000011efbe530, 2) = 0
1150978315.443078 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) = 0
1150978315.443591 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) = 0
1150978315.443928 [10000]{55517} #1 semop(34, 0xc00000011c6032b0, 1) = 0
1150978315.444411 [10000]{55517} #1 semop(16415, 0xc00000011efbe530, 2) = 0
1150978315.444838 [10000]{55517} #1 semop(16415, 0xc00000011efbe530, 2) = 0
1150978315.445309 [10000]{55517} #1 semop(16415, 0xc00000011f3d6530, 2) = 0
I├В┬┤m try undestand about the other link you post here.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
тАО06-26-2006 10:46 PM
тАО06-26-2006 10:46 PM
Re: SuperDome x Informix
This is output from my system:
u122:/# ipcs -mob
IPC status from /dev/kmem as of Tue Jun 27 07:23:10 2006
T ID KEY MODE OWNER GROUP NATTCH SEGSZ
Shared Memory:
m 0 0x411c1146 --rw-rw-rw- root root 0 348
m 1 0x4e0c0002 --rw-rw-rw- root root 1 61760
m 2 0x41201627 --rw-rw-rw- root root 1 8192
m 163843 0x06347849 --rw-rw-rw- root root 2 65544
m 98308 0x0c6629c9 --rw-r----- root root 2 34906344
m 131077 0x4914b152 --rw-r--rw- root root 0 22908
m 98310 0x00a5c581 --rw------- sfmdb users 1 10493952
m 17891335 0x52574801 --rw-rw---- root informix 50 4294705152
m 1376264 0x52574802 --rw-rw---- root informix 50 4294705152
m 1245193 0x52574803 --rw-rw---- root informix 50 2594529280
m 1114122 0x52574804 --rw-rw-rw- root informix 28 1155072