running onstat commands against a shared memory du
Posted in 2006
A user on IDS 10.00.FC5 under RHEL 4 couldn't run onstat against a ~4.3GB shared-memory dump (shmem.*.*) after an assertion failure, getting "rinfile: shmget/shmat error ... errno=28" (ENOSPC); IBM support claimed the technique was unsupported. Suggestions covered ulimit and 2GB filesystem limits (both ruled out). The explanation offered: onstat must attach the whole dump into shared memory, and with the live 4GB instance already running on the 6GB box there isn't room. Advice was to use a machine with plenty of memory/swap (slow), but no confirmed fix is recorded.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
Topics: Platform-Specific Issues, Versions, Editions & End-of-Life
OS:RH Linux 4.0 Patch 5 ES
Kernel: 2.6.9-42.ELsmp
IDS 10.00.FC5
6GB memory
2-CPU's
The database had a assertion failure last night while executing some
nightly processes.
I opened a case with IBM and sent them the assert file. Then I asked
the support person
why I was unable to issue onstat commands against the shmem.*.* dump
files like I was
shown in an Informix down systems class several years ago. I was told
that running
onstat commands against the shmem dump file was unsupported. Is it?
When I run onstat against the shmem dump file I receive the following error:
[informix@billdb1 tmp]$ onstat -d shmem.483dd89.0
rinfile: shmget/shmat error (cnt 3758096384) errno=28
Any ideas?
Thanks
Peter
All I know is on my box (AIX 5.2):
#define ENOSPC 28 /* No space left on device */
I used to work at Informix and they had us doing that command quite a bit
.....
Take care.
Clifton
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Peter
J Diaz de Leon
Sent: Thursday, September 21, 2006 11:32 AM
To: ids@iiug.org
Subject: running onstat commands against a shared memor.... [7513]
OS:RH Linux 4.0 Patch 5 ES
Kernel: 2.6.9-42.ELsmp
IDS 10.00.FC5
6GB memory
2-CPU's
The database had a assertion failure last night while executing some
nightly processes.
I opened a case with IBM and sent them the assert file. Then I asked
the support person
why I was unable to issue onstat commands against the shmem.*.* dump
files like I was
shown in an Informix down systems class several years ago. I was told
that running
onstat commands against the shmem dump file was unsupported. Is it?
When I run onstat against the shmem dump file I receive the following error:
[informix@billdb1 tmp]$ onstat -d shmem.483dd89.0
rinfile: shmget/shmat error (cnt 3758096384) errno=28
Any ideas?
Thanks
Peter
****************************************************************************
***
Forum Note: Use "Reply" to post a response in the discussion forum.
Hmmmm.
df -h tells me I have 17GB available on the partition where the shmem
dump files resides.
-Peter
Clifton Bean wrote:
> All I know is on my box (AIX 5.2):
> #define ENOSPC 28 /* No space left on device */
>
> I used to work at Informix and they had us doing that command quite a bit
> ......
>
> Take care.
> Clifton
>
> -----Original Message-----
> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of Peter
> J Diaz de Leon
> Sent: Thursday, September 21, 2006 11:32 AM
> To: ids@iiug.org
> Subject: running onstat commands against a shared memor.... [7513]
>
> OS:RH Linux 4.0 Patch 5 ES
> Kernel: 2.6.9-42.ELsmp
> IDS 10.00.FC5
> 6GB memory
> 2-CPU's
>
> The database had a assertion failure last night while executing some
> nightly processes.
> I opened a case with IBM and sent them the assert file. Then I asked
> the support person
> why I was unable to issue onstat commands against the shmem.*.* dump
> files like I was
> shown in an Informix down systems class several years ago. I was told
> that running
> onstat commands against the shmem dump file was unsupported. Is it?
>
> When I run onstat against the shmem dump file I receive the following error:
>
> [informix@billdb1 tmp]$ onstat -d shmem.483dd89.0
> rinfile: shmget/shmat error (cnt 3758096384) errno=28
>
> Any ideas?
>
> Thanks
> Peter
>
> ****************************************************************************
> ***
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
Yes but is that before, after, or during the time you run the command or
just after?
To really know if it's a space problem.... you have to see it when it
happens...
-----Original Message-----
From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
Peter J Diaz de Leon
Sent: Thursday, September 21, 2006 11:45 AM
To: ids@iiug.org
Subject: Re: running onstat commands against a shared m.... [7515]
Hmmmm.
df -h tells me I have 17GB available on the partition where the shmem
dump files resides.
-Peter
Clifton Bean wrote:
> All I know is on my box (AIX 5.2):
> #define ENOSPC 28 /* No space left on device */
>
> I used to work at Informix and they had us doing that command quite a
bit
> ......
>
> Take care.
> Clifton
>
> -----Original Message-----
> From: ids-bounces@iiug.org [mailto:ids-bounces@iiug.org] On Behalf Of
Peter
> J Diaz de Leon
> Sent: Thursday, September 21, 2006 11:32 AM
> To: ids@iiug.org
> Subject: running onstat commands against a shared memor.... [7513]
>
> OS:RH Linux 4.0 Patch 5 ES
> Kernel: 2.6.9-42.ELsmp
> IDS 10.00.FC5
> 6GB memory
> 2-CPU's
>
> The database had a assertion failure last night while executing some
> nightly processes.
> I opened a case with IBM and sent them the assert file. Then I asked
> the support person
> why I was unable to issue onstat commands against the shmem.*.* dump
> files like I was
> shown in an Informix down systems class several years ago. I was told
> that running
> onstat commands against the shmem dump file was unsupported. Is it?
>
> When I run onstat against the shmem dump file I receive the following
error:
>
> [informix@billdb1 tmp]$ onstat -d shmem.483dd89.0
> rinfile: shmget/shmat error (cnt 3758096384) errno=28
>
> Any ideas?
>
> Thanks
> Peter
>
>
************************************************************************
****
> ***
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
>
************************************************************************
*******
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
************************************************************************
*******
Forum Note: Use "Reply" to post a response in the discussion forum.
============================================================
The information contained in this message may be privileged
and confidential and protected from disclosure. If the reader
of this message is not the intended recipient, or an employee
or agent responsible for delivering this message to the
intended recipient, you are hereby notified that any reproduction,
dissemination or distribution of this communication is strictly
prohibited. If you have received this communication in error,
please notify us immediately by replying to the message and
deleting it from your computer. Thank you. Tellabs
============================================================
Two Questions,
#1 What is your ulimit set to?
IN most cases the default is around 200 MB, is that smaller than the size of
your shared memory.
#2 I'm going to admit to a little Linux ignorance here, but on most other OS's
they have at least one file system that only allows a maximum file size of
2Gig, and that is usually the default file system. I know for example that you
need to set AIX to JFS2 in order to get larger than 2 Gig files on a file
system.
----- Original Message ----
From: Peter J Diaz de Leon <pdiazdeleon@infinityhealthcare.com>
To: ids@iiug.org
Sent: Thursday, September 21, 2006 11:32:17 AM
Subject: running onstat commands against a shared memor.... [7513]
OS:RH Linux 4.0 Patch 5 ES
Kernel: 2.6.9-42.ELsmp
IDS 10.00.FC5
6GB memory
2-CPU's
The database had a assertion failure last night while executing some
nightly processes.
I opened a case with IBM and sent them the assert file. Then I asked
the support person
why I was unable to issue onstat commands against the shmem.*.* dump
files like I was
shown in an Informix down systems class several years ago. I was told
that running
onstat commands against the shmem dump file was unsupported. Is it?
When I run onstat against the shmem dump file I receive the following error:
[informix@billdb1 tmp]$ onstat -d shmem.483dd89.0
rinfile: shmget/shmat error (cnt 3758096384) errno=28
Any ideas?
Thanks
Peter
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
The ulimit is set to "unlimited."
The the two shared memory dumps are both 4317282304 in size.
Well over any 2GB limit. The file system in uses is EXT3.
Ideas?
Thanks
Peter
Mark Jamison wrote:
> Two Questions,
>
> #1 What is your ulimit set to?
> IN most cases the default is around 200 MB, is that smaller than the size of
> your shared memory.
>
> #2 I'm going to admit to a little Linux ignorance here, but on most other
OS's
> they have at least one file system that only allows a maximum file size of
> 2Gig, and that is usually the default file system. I know for example that
you
> need to set AIX to JFS2 in order to get larger than 2 Gig files on a file
> system.
>
> ----- Original Message ----
> From: Peter J Diaz de Leon <pdiazdeleon@infinityhealthcare.com>
> To: ids@iiug.org
> Sent: Thursday, September 21, 2006 11:32:17 AM
> Subject: running onstat commands against a shared memor.... [7513]
>
> OS:RH Linux 4.0 Patch 5 ES
> Kernel: 2.6.9-42.ELsmp
> IDS 10.00.FC5
> 6GB memory
> 2-CPU's
>
> The database had a assertion failure last night while executing some
> nightly processes.
> I opened a case with IBM and sent them the assert file. Then I asked
> the support person
> why I was unable to issue onstat commands against the shmem.*.* dump
> files like I was
> shown in an Informix down systems class several years ago. I was told
> that running
> onstat commands against the shmem dump file was unsupported. Is it?
>
> When I run onstat against the shmem dump file I receive the following error:
>
> [informix@billdb1 tmp]$ onstat -d shmem.483dd89.0
> rinfile: shmget/shmat error (cnt 3758096384) errno=28
>
> Any ideas?
>
> Thanks
> Peter
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
Is it on the same box as the instance itself?
Because onstat against a dump file has to load the entire file into memory.
So if you have 6 Gig, and 4 Gig is already taken by the informix instance, you
will get error -28 attempting to load the segment.
----- Original Message ----
From: Peter J Diaz de Leon <pdiazdeleon@infinityhealthcare.com>
To: ids@iiug.org
Sent: Thursday, September 21, 2006 1:38:36 PM
Subject: Re: running onstat commands against a shared m.... [7518]
The ulimit is set to "unlimited."
The the two shared memory dumps are both 4317282304 in size.
Well over any 2GB limit. The file system in uses is EXT3.
Ideas?
Thanks
Peter
Mark Jamison wrote:
> Two Questions,
>
> #1 What is your ulimit set to?
> IN most cases the default is around 200 MB, is that smaller than the size of
> your shared memory.
>
> #2 I'm going to admit to a little Linux ignorance here, but on most other
OS's
> they have at least one file system that only allows a maximum file size of
> 2Gig, and that is usually the default file system. I know for example that
you
> need to set AIX to JFS2 in order to get larger than 2 Gig files on a file
> system.
>
> ----- Original Message ----
> From: Peter J Diaz de Leon <pdiazdeleon@infinityhealthcare.com>
> To: ids@iiug.org
> Sent: Thursday, September 21, 2006 11:32:17 AM
> Subject: running onstat commands against a shared memor.... [7513]
>
> OS:RH Linux 4.0 Patch 5 ES
> Kernel: 2.6.9-42.ELsmp
> IDS 10.00.FC5
> 6GB memory
> 2-CPU's
>
> The database had a assertion failure last night while executing some
> nightly processes.
> I opened a case with IBM and sent them the assert file. Then I asked
> the support person
> why I was unable to issue onstat commands against the shmem.*.* dump
> files like I was
> shown in an Informix down systems class several years ago. I was told
> that running
> onstat commands against the shmem dump file was unsupported. Is it?
>
> When I run onstat against the shmem dump file I receive the following error:
>
> [informix@billdb1 tmp]$ onstat -d shmem.483dd89.0
> rinfile: shmget/shmat error (cnt 3758096384) errno=28
>
> Any ideas?
>
> Thanks
> Peter
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
Yeah it's on the same box. It the only box currently with 6GB on it.
Is there anyway to run the onstat on a box with less memory?
-Peter
Mark Jamison wrote:
>Is it on the same box as the instance itself?
>
>Because onstat against a dump file has to load the entire file into memory.
>
>So if you have 6 Gig, and 4 Gig is already taken by the informix instance, you
>will get error -28 attempting to load the segment.
>
>----- Original Message ----
>From: Peter J Diaz de Leon <pdiazdeleon@infinityhealthcare.com>
>To: ids@iiug.org
>Sent: Thursday, September 21, 2006 1:38:36 PM
>Subject: Re: running onstat commands against a shared m.... [7518]
>
>The ulimit is set to "unlimited."
>
>The the two shared memory dumps are both 4317282304 in size.
>Well over any 2GB limit. The file system in uses is EXT3.
>
>Ideas?
>Thanks
>Peter
>
>Mark Jamison wrote:
>
>
>>Two Questions,
>>
>>#1 What is your ulimit set to?
>>IN most cases the default is around 200 MB, is that smaller than the size of
>>your shared memory.
>>
>>#2 I'm going to admit to a little Linux ignorance here, but on most other
>>
>>
>OS's
>
>
>>they have at least one file system that only allows a maximum file size of
>>2Gig, and that is usually the default file system. I know for example that
>>
>>
>you
>
>
>>need to set AIX to JFS2 in order to get larger than 2 Gig files on a file
>>system.
>>
>>----- Original Message ----
>>From: Peter J Diaz de Leon <pdiazdeleon@infinityhealthcare.com>
>>To: ids@iiug.org
>>Sent: Thursday, September 21, 2006 11:32:17 AM
>>Subject: running onstat commands against a shared memor.... [7513]
>>
>>OS:RH Linux 4.0 Patch 5 ES
>>Kernel: 2.6.9-42.ELsmp
>>IDS 10.00.FC5
>>6GB memory
>>2-CPU's
>>
>>The database had a assertion failure last night while executing some
>>nightly processes.
>>I opened a case with IBM and sent them the assert file. Then I asked
>>the support person
>>why I was unable to issue onstat commands against the shmem.*.* dump
>>files like I was
>>shown in an Informix down systems class several years ago. I was told
>>that running
>>onstat commands against the shmem dump file was unsupported. Is it?
>>
>>When I run onstat against the shmem dump file I receive the following error:
>>
>>[informix@billdb1 tmp]$ onstat -d shmem.483dd89.0
>>rinfile: shmget/shmat error (cnt 3758096384) errno=28
>>
>>Any ideas?
>>
>>Thanks
>>Peter
>>
>>
>>
On a box with lots of virtual memory (swap) but it will be VEERRRYYYY
SSSLLLOOOWWW.
Art S. Kagel
----- Original Message -----
From: Peter J Diaz De Leon <ids@iiug.org>
At: 9/22 5:42:59
Yeah it's on the same box. It the only box currently with 6GB on it.
Is there anyway to run the onstat on a box with less memory?
-Peter
Mark Jamison wrote:
>Is it on the same box as the instance itself?
>
>Because onstat against a dump file has to load the entire file into memory.
>
>So if you have 6 Gig, and 4 Gig is already taken by the informix instance,
you
>will get error -28 attempting to load the segment.
>
>----- Original Message ----
>From: Peter J Diaz de Leon <pdiazdeleon@infinityhealthcare.com>
>To: ids@iiug.org
>Sent: Thursday, September 21, 2006 1:38:36 PM
>Subject: Re: running onstat commands against a shared m.... [7518]
>
>The ulimit is set to "unlimited."
>
>The the two shared memory dumps are both 4317282304 in size.
>Well over any 2GB limit. The file system in uses is EXT3.
>
>Ideas?
>Thanks
>Peter
>
>Mark Jamison wrote:
>
>
>>Two Questions,
>>
>>#1 What is your ulimit set to?
>>IN most cases the default is around 200 MB, is that smaller than the size of
>>your shared memory.
>>
>>#2 I'm going to admit to a little Linux ignorance here, but on most other
>>
>>
>OS's
>
>
>>they have at least one file system that only allows a maximum file size of
>>2Gig, and that is usually the default file system. I know for example that
>>
>>
>you
>
>
>>need to set AIX to JFS2 in order to get larger than 2 Gig files on a file
>>system.
>>
>>----- Original Message ----
>>From: Peter J Diaz de Leon <pdiazdeleon@infinityhealthcare.com>
>>To: ids@iiug.org
>>Sent: Thursday, September 21, 2006 11:32:17 AM
>>Subject: running onstat commands against a shared memor.... [7513]
>>
>>OS:RH Linux 4.0 Patch 5 ES
>>Kernel: 2.6.9-42.ELsmp
>>IDS 10.00.FC5
>>6GB memory
>>2-CPU's
>>
>>The database had a assertion failure last night while executing some
>>nightly processes.
>>I opened a case with IBM and sent them the assert file. Then I asked
>>the support person
>>why I was unable to issue onstat commands against the shmem.*.* dump
>>files like I was
>>shown in an Informix down systems class several years ago. I was told
>>that running
>>onstat commands against the shmem dump file was unsupported. Is it?
>>
>>When I run onstat against the shmem dump file I receive the following error:
>>
>>[informix@billdb1 tmp]$ onstat -d shmem.483dd89.0
>>rinfile: shmget/shmat error (cnt 3758096384) errno=28
>>
>>Any ideas?
>>
>>Thanks
>>Peter
>>
>>
>>
*******************************************************************************
Forum Note: Use "Reply" to post a response in the discussion forum.
Related threads
- IDS 10 table-level restore
- Informix Development Webinar December 11, 2007
- ontape -p/r with changed ROOTPATH
- Migrate from HP PA-RISC to HP ITANIUM by ontape