| Author |
Message |
David
Guest
|
Posted:
Wed May 11, 2005 4:54 pm Post subject:
SNA Manage Agent with Event 5123 & 5150 |
|
|
I would like to know if anyone have any idea what I can do about a
problem that happen once a month ? I got these Events a lot in the app
log and the only thing that seems to work is to reboot the server.
Event Type: Error
Event Source: SNA Manage Agent
Event Category: None
Event ID: 5123
Date: 5/6/2005
Time: 2:22:10 PM
User: N/A
Computer: Server
Description:
Virtual allocation for descriptor has failed.
Event Type: Error
Event Source: SNA Manage Agent
Event Category: None
Event ID: 5150
Date: 5/6/2005
Time: 2:22:10 PM
User: N/A
Computer: Server
Description:
Manage base error in list processing.Increase queue size failed. Size
34816. Line 704.
If it can help, I have HIS 2000 SP1 on a W2K server SP4.
Thanks.
David |
|
| Back to top |
|
 |
Neil Pike
Guest
|
Posted:
Wed May 11, 2005 11:48 pm Post subject:
Re: SNA Manage Agent with Event 5123 & 5150 |
|
|
David,
That does sound like a memory/resource leak. I'd say report it to PSS, but
first the usual advice of updating to the latest service-pack - SP2 for HIS2000
was released not so long ago. I stuck it on my HIS 2000 machines with no
problems so far
Neil Pike. Protech Computing Ltd |
|
| Back to top |
|
 |
Stephen Jackson [MSFT]
Guest
|
Posted:
Mon May 16, 2005 4:52 pm Post subject:
Re: SNA Manage Agent with Event 5123 & 5150 |
|
|
David,
The Event 5123 and Event 5150 have been seen occasionally at customer sites.
Typically, the events go away when the MngAgent service is restarted and do
not reappear or take a long time to reappear. The following is meant to
explain one scenario that can result in these events being logged and it
also explains how status updates are handled in SNA Manager.
When you select an object (a Local APPC LU, for example) in the left panel
of the SNA Manager MMC snap-in, a register notification is send down to
mngcli.exe to start to send status information for the object. Mngcli will
then put any changes into a notification queue (implemented in mngcli's
memory by mngbase). SNA Manager then starts to pull them from the queue and
update the list view in the right pane of the SNA Manager window. If the
status changesfor the selected object are of a signification number (occurs
rapidly), the number of entries in the notification queue will increase
quickly. If you select the active users folder, the local LU object or the
remote LU object, all of these changes need to be updated in the list view.
It appears that the SNA Manager is not able to read them out of the queue
fast enough, so we end up with more and more sessions listed. Eventually the
pool for this queue will be max'ed out and mngcli will start to log the
events 5123 and 5150.
Do you know if SNA Manager is open on this server and if perhaps something
like the Print Service (or anythign that has a large number of status
changes) is selected in the left-hand pane of SNA Manager?
The options at this point are as follows:
1) Restart the MngAgent service (this will not affect any users or
sessions).
OR
2) Try to gather more diagnostics to determine the cause of the problem and
open a support case with Microsoft PSS. These diagnostics would include the
following:
- Manage Client, Manage Agent, and Manger internal traces
- Event logs (system and application)
Depending on what these show, we may need to gather some other diagnostics
as well.
--
Stephen Jackson
Microsoft® SNA/HIS Support
Please do not send e-mail directly to this alias. This alias is for
newsgroup purposes only. This posting is provided "AS IS"
with no warranties, and confers no rights.
"Neil Pike" <neilpike@compuserve.com> wrote in message
news:VA.000062e1.024ad470@compuserve.com...
| Quote: | David,
That does sound like a memory/resource leak. I'd say report it to PSS,
but
first the usual advice of updating to the latest service-pack - SP2 for
HIS2000
was released not so long ago. I stuck it on my HIS 2000 machines with no
problems so far
Neil Pike. Protech Computing Ltd
|
|
|
| Back to top |
|
 |
L Ring
Guest
|
Posted:
Mon Jun 13, 2005 12:52 am Post subject:
Re: SNA Manage Agent with Event 5123 & 5150 |
|
|
I concur. I had this happen alot and it was related to the MMC SNA Management
console. Eventually this cleared up for us. I will check but I ran this by
PSS awhile back and in our case it had something to do with the way I did our
upgrade coupled with making certain changes affecting connections and pus
without stopping the SNA Service on the affected server.
"Stephen Jackson [MSFT]" wrote:
| Quote: | David,
The Event 5123 and Event 5150 have been seen occasionally at customer sites.
Typically, the events go away when the MngAgent service is restarted and do
not reappear or take a long time to reappear. The following is meant to
explain one scenario that can result in these events being logged and it
also explains how status updates are handled in SNA Manager.
When you select an object (a Local APPC LU, for example) in the left panel
of the SNA Manager MMC snap-in, a register notification is send down to
mngcli.exe to start to send status information for the object. Mngcli will
then put any changes into a notification queue (implemented in mngcli's
memory by mngbase). SNA Manager then starts to pull them from the queue and
update the list view in the right pane of the SNA Manager window. If the
status changesfor the selected object are of a signification number (occurs
rapidly), the number of entries in the notification queue will increase
quickly. If you select the active users folder, the local LU object or the
remote LU object, all of these changes need to be updated in the list view.
It appears that the SNA Manager is not able to read them out of the queue
fast enough, so we end up with more and more sessions listed. Eventually the
pool for this queue will be max'ed out and mngcli will start to log the
events 5123 and 5150.
Do you know if SNA Manager is open on this server and if perhaps something
like the Print Service (or anythign that has a large number of status
changes) is selected in the left-hand pane of SNA Manager?
The options at this point are as follows:
1) Restart the MngAgent service (this will not affect any users or
sessions).
OR
2) Try to gather more diagnostics to determine the cause of the problem and
open a support case with Microsoft PSS. These diagnostics would include the
following:
- Manage Client, Manage Agent, and Manger internal traces
- Event logs (system and application)
Depending on what these show, we may need to gather some other diagnostics
as well.
--
Stephen Jackson
Microsoft® SNA/HIS Support
Please do not send e-mail directly to this alias. This alias is for
newsgroup purposes only. This posting is provided "AS IS"
with no warranties, and confers no rights.
"Neil Pike" <neilpike@compuserve.com> wrote in message
news:VA.000062e1.024ad470@compuserve.com...
David,
That does sound like a memory/resource leak. I'd say report it to PSS,
but
first the usual advice of updating to the latest service-pack - SP2 for
HIS2000
was released not so long ago. I stuck it on my HIS 2000 machines with no
problems so far
Neil Pike. Protech Computing Ltd
|
|
|
| Back to top |
|
 |
|
|
|
|