Roy,
I have had a quick look at this on Ingres 2006 r2 (windows) and the
classids used in ima_cssampler_stats do not exist.
The table registration looks like this:
register table ima_cssampler_stats (
server varchar(32) not null not default is 'SERVER',
samp_index varchar(32) not null not default is
'exp.clf.nt.cs.samp_index',
numsamples integer4 not null not default is
'exp.clf.nt.cs.samp_numsamples',
numtlssamples integer4 not null not default is
'exp.clf.nt.cs.samp_numtlssamples',
numtlsslots integer4 not null not default is
'exp.clf.nt.cs.samp_numtlsslots',
numtlsprobes integer4 not null not default is
'exp.clf.nt.cs.samp_numtlsprobes',
numtlsreads integer4 not null not default is
'exp.clf.nt.cs.samp_numtlsreads',
numtlsdirty integer4 not null not default is
'exp.clf.nt.cs.samp_numtlsdirty',
numtlswrites integer4 not null not default is
'exp.clf.nt.cs.samp_numtlswrites'
)
as import from 'tables'
with dbms = IMA,
structure = sortkeyed,
key = (server);
If you:
Select distinct classid from ima_mib_objects where classid like
'exp.clf.nt.cs.samp%'
No rows are returned, a sure sign that the classids do not exist.
If you:
Select distinct classid from ima_mib_objects where classid like
'exp.clf.nt%'
Note: if you are on unix or linux then it is 'exp.clf.unix%' and for
vms it is 'exp.clf.vms%'
I got back 149 rows which includes such nice classids as:
exp.clf.nt.cs.srv_block.wait_stats.dior_time
exp.clf.nt.cs.srv_block.wait_stats.dior_waits
exp.clf.nt.cs.srv_block.wait_stats.diow_done
exp.clf.nt.cs.srv_block.wait_stats.diow_time
exp.clf.nt.cs.srv_block.wait_stats.diow_waits
exp.clf.nt.cs.srv_block.wait_stats.lg_done
exp.clf.nt.cs.srv_block.wait_stats.lg_idle
exp.clf.nt.cs.srv_block.wait_stats.lg_time
exp.clf.nt.cs.srv_block.wait_stats.lg_waits
exp.clf.nt.cs.srv_block.wait_stats.lge_done
exp.clf.nt.cs.srv_block.wait_stats.lge_time
exp.clf.nt.cs.srv_block.wait_stats.lge_waits
exp.clf.nt.cs.srv_block.wait_stats.lio_done
exp.clf.nt.cs.srv_block.wait_stats.lio_time
exp.clf.nt.cs.srv_block.wait_stats.lio_waits
exp.clf.nt.cs.srv_block.wait_stats.lior_done
exp.clf.nt.cs.srv_block.wait_stats.lior_time
exp.clf.nt.cs.srv_block.wait_stats.lior_waits
exp.clf.nt.cs.srv_block.wait_stats.liow_done
exp.clf.nt.cs.srv_block.wait_stats.liow_time
exp.clf.nt.cs.srv_block.wait_stats.liow_waits
exp.clf.nt.cs.srv_block.wait_stats.lk_done
"DBMS Sampling" output has changed and my guess is that the MIB objects
were changed and the output changed but nobody changed IMA (hum that
sounds familiar)
What to do:
Log it as a BUG with Support and wait
Or better still create new IMA tables to reflect the new "DBMS sampling"
MIB Objects and submit them as part of "Ingres Janitors"
Cheers
John Smedley
-----Original Message-----
From: info-ingres-bounces@kettleriverconsulting.com
[mailto:info-ingres-bounces@kettleriverconsulting.com] On Behalf Of Roy
Hann
Sent: 26 November 2007 14:14
To: info-ingres@kettleriverconsulting.com
Subject: [Info-Ingres] querying ima_cssampler_stats
Has anyone had any joy getting data out of ima_cssampler_stats in imadb?
I
notice there's an ima_start_sampler procedure but neither that nor
starting
sampling manually with iimonitor seems to produce any results in
ima_cssampler_stats.
Roy
_______________________________________________
Info-Ingres mailing list
Info-Ingres@kettleriverconsulting.com
http://www.kettleriverconsulting.com/mailman/listinfo/info-ingres
Roy Hann - 27 Nov 2007 07:25 GMT
> Roy,
>
> I have had a quick look at this on Ingres 2006 r2 (windows) and the
> classids used in ima_cssampler_stats do not exist.
[snip]
> What to do:
>
> Log it as a BUG with Support and wait
>
> Or better still create new IMA tables to reflect the new "DBMS sampling"
> MIB Objects and submit them as part of "Ingres Janitors"
Thanks John. I reckon I'm pretty much duty-bound to do just that. I might
even go looking for other gaps.
Roy
withdefault@gmail.com - 27 Nov 2007 10:27 GMT
On Nov 27, 7:25 am, "Roy Hann" <specia...@processed.almost.meat>
wrote:
> > Roy,
>
[quoted text clipped - 14 lines]
>
> Roy
Roy,
I've taken a look at the Ingres 2006 r2 code and it seems that the
class ids are defined on the Windows platform only. This is
definitely a bug/omission for Linux and other platforms.
These are some of the few IMA objects that are platform dependent and
the IMA database set-up script for UNIX still references
exp.clf.nt.... in the table registrations, so even if the table
registration is fixed the objects values still won't appear.
Ray
ingres1990-groups@yahoo.co.uk - 27 Nov 2007 15:04 GMT
On 27 Nov, 10:27, withdefa...@gmail.com wrote:
> On Nov 27, 7:25 am, "Roy Hann" <specia...@processed.almost.meat>
> wrote:
[quoted text clipped - 34 lines]
>
> Ray
Ray,
It does not work on Windows either
Steve
withdefault@gmail.com - 27 Nov 2007 18:06 GMT
On Nov 27, 3:04 pm, ingres1990-gro...@yahoo.co.uk wrote:
> On 27 Nov, 10:27, withdefa...@gmail.com wrote:
>
[quoted text clipped - 42 lines]
>
> Steve
Steve,
I just tried it on Ingres 2006 Release 2 Microsoft Windows 32-bit
Version II 9.1.0 (int.w32/117) and it works for me. You will only get
results in the ima_cssampler_stats table whilst the sampler is
started. The table is empty after the sampling is stopped.
Ray
ingres1990-groups@yahoo.co.uk - 28 Nov 2007 09:49 GMT
On 27 Nov, 18:06, withdefa...@gmail.com wrote:
> On Nov 27, 3:04 pm, ingres1990-gro...@yahoo.co.uk wrote:
>
[quoted text clipped - 53 lines]
>
> Ray
Ray,
Thanks
You are quite right - I thought I had tried it with sampling on -
obviously I didn't.
Steve
Yehudah Edelstein - 28 Nov 2007 16:35 GMT
Hi,
We have an old Ingres system that exists for retrieval only,
almost 7 years.
We have
Ingres Alpha VMS Version II 2.0/0308 (axm.vms/00)
>From the days CA gave the support.
It runs on
OpenVMS V7.3-1 .
The upgrade of our central storage system, requires we upgrade to
OpenVMS V7.3-2.
Anyone have experience with this? Can this upgrade be done with no
problems? Also the storage disks will change.
Who supports Ingres today?
Thanks for any advice,
Yehudah
---------------------------------------------------------------------------------------------------------------
This e-mail message may contain confidential, commercial and privileged information or data that
constitute proprietary information of Cellcom Israel Ltd. Any review or distribution by others is
strictly prohibited. If you are not the intended recipient you are hereby notified that any use
of this information or data by any other person is absolutely prohibited.
If you are not the intended recipient, please delete all copies.
Thank You.
http://www.cellcom.co.il
==============================================================================