XI if SLD is down

Hello,
what happens to XI processes when SLD is down ? SLD being in cache, I think XI should continue to work, isn't it ? If yes, what happens when there is a cache refresh ?
Thanks in advance.

Hi Elliot,
As design and configuration depends on SLD.
So when SLD is down then XI may work untill cache get refreshed.
once SLD chache get refreshed then we will find error in XI.
chache refresh are mainly of two type
automatically and manually
there are setting to refresh SLD chache after some time.
in dirty chache refresh one can work with XI while cotinuously.
So XI can work till SLD chache get refreshed.
Please reward points if beneficial.
please revert if further query
Regards,
Sumit Gupta
[email protected]

Similar Messages

If my SLD is down will webdynpro apps fail.

I see that the SLD manages the JCO connections to R/3. Does the engine maintain that data locally as well or if the SLD is down then the webdynpro app utilizing the JCO connection is down as well.

Unfortunately the answer is no the information is not cached and yes it will fail if the SLD is not running.

PI problems if SLD Shut down ?

Hi,
I'm currious, ou PI is setup to use its local SLD. This local SLD receives updates through full synchronisation from a central SLD.
Now , if we shut down, the central SLD. Will everything still work in PI ? Exp : transports... etc.
Obviously , I can't transport new software components and products but what about bug fixes to existing software components ?
Thank you.
Edited by: Thierry Dagnino on Nov 4, 2009 5:16 PM

Export/ import (synchronization) done from the Central SLD into the local SLD done before shutdown should help local SLD to work properly in case the central instance goes down. Have not tried practically....above statement is based on my reading....if you feel its wrong then feel free to correct it.
Regards,
Abhishek.

SLD down but cache refresh

Hi,
If the SLD is down (unavailable) and we make a cache refresh. What happen ? Should the refresh fail or everything will be erased ?
Thx.

hi
As far as i know....
if the sld is down ur cache refresh will fail....
actually whenever we refresh cache ..it will access the SLD ( to get the details of entegration engine ,adapter engine etc.....)....So if no connection is there then cache refresh will not work......
Bill
**reward points if it helps you

SLD - Data Suppliers

I have recently updated our solution Manager system to SPS22 with java.
Since then the Data Suppliers in the SLD are DOWN.
Administration > Details > Data suppliers
RFC server threads:
[DOWN] NUC:<server>_<port>_SLD_NUC_1448281814
[DOWN] UC ::<server>_<port>__SLD_UC_1102336995
[DOWN] NUC::<server>_<port>__SLD_NUC_2003281767
[DOWN] UC ::<server>_<port>__SLD_UC_1129708313
Can anybody tell me if there is a problem with the latest version of Java or tell me how to get the data suppliers back up?
Also the Landscape Fetch job now fails with The SLD server connection is inactive. Is this a consiquence of the data suppliers?
Hope you can help.
Russell Burrow

Hello,
Hope the SLD server is in Running status!
In ABAP stack, is the t-code SLDCHECK for the corresponding system working fine, all GREEN? If not working, check the RFC connection issues. The Landscape Fetch job would be failing due to the same reason.
In Java stack, in VA, check the SLD Datasupplier is working or not. Also check the CIM connectivity to SLD server. if not working, check what is the error.
Hope this helps.
Regards,
Bhaskar

Best Practise for multiple SLDs and CMS

Hi Guys,
 We are planning on our XI Landscape, can you guide me to best practices document or weblogs on where it answers my following points.
We are planning to have DEV - QA - PROD XI Boxes. We are planning to implement CMS. What is the optimal approach, pros and cons of each approach.
1) Installing one central SLD which would be either on DEV or QA box. CMS also installed on the same box.
2) Installing two SLDs
 i) one on Dev Box. CMS on Dev Box sharing SLD
 ii) second SLD on QA box which is shared by PROD box which will be the SLD used for production.
I would be happy to see responses with any existing landscapes which are being implemented by you guys in case of CMS also in picture.
Thanks in Advance
Regards
Sujan

Hi,
In my project we have maitained 2 SLD one for Production landscape and one for non-Production landscape. However the High Availability scenario is available in 7.1 but it is not there in release 7.0.
In case of release 7.0 you have to follow the following procedure to have the similar (HA) kind of scenario:
1). In case of ABAP and dual stack go to SLDAPICUST and maitain two entries for SLD with proper host name, port number and PIAAPLUSER detail.
Make one entry as Primary.
Also if JAVA stack is also there then go to Admin tool and maintain the detail in "HTTP Setting" of the "SLD Data Service" with the data of the primary SLD just maintained above through SLDAPICUST.
2). In case of only JAVA stack maitain the detail in "HTTP Setting" of the "SLD Data Service" with the data of the primary SLD.
Note--Here there is no option to maintain two SLD server details.
3). Now Login to ABAP/+JAVA stack server at the SAP GUI level and check the connection by executing SLDCHECK command. If it works fine then do the following:
 a). Now to make your two SLD in sync login to SLD with Admin user having write access to LCR (e.g. PISUPER, j2ee_admin in case of ABAP+JAVA or "administrator" user in case of JAVA stack) and go to Administration->Data Suppliers->Add SLD.
Here you have to add the other SLD detail in the form of "http://<servername>:<j2eeport>/, with SLDDSUSER ID detail. This will make both the SLD in sync to each other.
Note--Make sure that the password for all the user mentioned above is same in both the SLD otherwise it would cause an error.
 b). Also take the full export of your primary SLD by going through "Administartion->Export" and Import it in your secondary SLD. This will make both the SLD in sync with each other initialy and you won't get any error when both SLD tried to sync automaticaly.
4). In the case when one SLD is down then make the other SLD primary through the transaction "SLDAPICUST", RZ70 and by logging through Admin tool. Check your setting through the transaction SLDCHECK.
Note--In RZ70 you have only one option and you can maitain only one SLD detail at a time. So whenever there is a problem in one SLD make sure that you have make the changes in RZ70 as well with the new detail of the SLD.
Best regards,
sheo

JCo connection and SLD

Greetings, we have a federated portal which is using JCo connections as part of the MSS/ESS business package. We had a situation where the SLD was down therefore the JCo connections were down. Our question is JCo connection completely dependent on SLD. In other words if SLD is down, will that mean our JCo connection will fail thus an outage.
If this is the case what is the best way to make sure SLD is highly available? Our options are to make SLD highly available or distribute SLD on each system.
Also, what is the performance hit on making SLD distributed on all Portal system if this the route we go?
Thanks in advance!

Hi,
The distributed SLD servers can cause higher administration expense with regard to consistent data. Therefore, we recommend that you consider to use a single SLD server first.
but for large system landscapes that are distributed over different geographic locations, a single SLD server might slow down the performance of involved application systems. In this case, a distributed SLDserver installation can be taken into account.
Samrat

Local SLD -Missing Jco information

Hi All,
I have a question with regards to our scenario..We have a 4 system lanscape with SND,DEV,QAS,PRD sytems and Solman 4.0. We have the central SLD setup in the solman We are currently having ESS in the SND portal which is installed on a stand alone Java stack. So if the Solman is down, the ESS in the SND portal does'nt work. So i setup a SLD bridge between Solman - SLD and the SND -SLD with the hope that the ESS would work even if the solman goes down. I can now see all the systems and the corresponding information (under technical systems ) in the SND-SLD, as can be seen in the SOLMAN-SLD. Only important thing that is missing is the Jco connection information in the SND-SLD, which can be seen in the SOLMAN -SLD.
Now whenever a user accesses the ESS application in the SND portal, it takes the JCo destination information from the central SLD which is in the Solman. Is it posssible that in case the Solman -SLD goes down, the local SND-SLD can help the ESS application in the SND portal to locate the Jco destination. (Because in the Webdynpro Content Administrator of the SND portal in the " Check SLD Connection " i can see information for the Solman -SLD .) What additional configuration needs to be done so that i can include SND-SLD information as well along with the SOLMAN-SLD.
Please suggest.....
Regards,

Hello Sai,
Thanks for the reply...yes i am aware that i need to change the settings in the SLD Data Supplier service in Visual admin on SND system so that it points to itself.But i am unsure if the ESS/MSS would work without the JCo destination information available in the SND-SLD.As i previously mentioned, i have set up the SND-SLD to receive the data from the Solman-SLD via the SLD Bridge.I can see all the systems and their information in the SND-SLD but i cannot the the Jco destinations in the SND-SLD for the portal systems.since the Jco information is not available in the SND-SLD the ESS/MSS in the SND portal would fail.....
Regards,
Mangesh

SLD Crazy

I could not find an SLD forum so this place seems as good as any.
An ABAP system has RFC's defined in each system and this has never been an issue.
Now we have J2EE's with adaptive RFC's defined in the SLD rather than in the local system. These 'SM59' style connections are not transferrable between systems as far as I can see and have little value in the SLD as opposed to the local system.
What I see most crazy of all is that after 2 hours these entries are deleted from the local cache and then fetched again fom the SLD. Surely if the fetch fails they should keep using the cached version?
It seems mad to me that they are stored in the SLD and that they time out before they are re-fetched therefore creating significant risk to productive system availability.
I now recommend that Portals have a local SLD that is never patched and have manually defined adaptive RFC connections. The SLD Provider service reads and writes to itself but bridges to a central SLD. These never seem to have issues and now no longer go down due to reliance on that SLD system and network between.
I am very well versed in the SLD and would appreciate responses from people who can correct me in my ways as SAP seems to require me to have a central SLD for these even though I have no maintenance and the GUIDs are not required elsewhere.

Hello Experts,
I am trying a similar scenario, I am already having a central SLD with Solution Manager 7.01 [Netweaver Release 7.0] Dual Stack and have activated [by initial setup] a local SLD based on Java Stack Only [Netweaver Release 7.0]
I am trying to figure out a possible way, once the central SLD and local SLD's are synced, if central SLD goes down the local SLD takes over from the Central SLD, would like to have a automated switching between the Central and Local SLD.
Again the frustrating thing is that I am unable to find an official SAP document explaining such a scenario, would appreciate any advice on this particular issue.
Regards
Umar Tariq

SLD and PI

Dear Gurus.
My question is about a role of SLD on PI work.
Does the PI need the SLD server available during runtime, not only at design phase?
For example PI using central SLD, or SLD running on another instance. In case that SLD is down - can I stop PI runtime?
Regards
Vladimir

Hi,
Dev, QA and Prod. are involved in flow of transport requests and this flow is enabled through a workflow process. Since all these flows towards Prod., hence it is clear that Prod. is the system where everything flows. However sometimes we also have the reverse flow (sometimes in case of idocs). So, considering the above scenario if we talk about master SLD, since we do not have any direct flow from dev. to Prod, or from Prod. to dev. hence the intermediate QA should be the master SLD since it directs flows towards various systems.
Regards,
Souvik

RFC Destination Between R/3 and XI Not Reestablishing

I have an RFC destination set up in R/3 so that calls can be made to our XI system. Everything works fine until our central SLD goes down for maintenance,etc. When the SLD is down, the RFC connection is lost. I assume this is by design. The problem is that once the SLD comes back up, I would expect the RFC Destination connection to automatically re-establish itself. It does not. I have to go into IB:Config and change/save/reactivate a channel using the RFC destination so that it gets reregistered.
Is there a setting that will make the RFC Destination connectivity re-establish itself automatically?
Thanks

When you say "go add some spaces" are you referring to putting spaces in the description field just so you can change/activate? I assume you did, just checking.
Thanks Naveen.
Everyone else out there: Do you all have to manually do this every time the SLD goes down?

Error in Mail Sender Adapter

Hi all,
without any obvious reason i'm suddenly getting an error in the runtime workbench at the Communication Channel Monitoring:
exception caught during processing mail message; java.io.IOException: unexpected greeting response; read * BYE Connection refused
I already restarted the j2ee engine, but with no success.
Can anybody please help me?
Points will be rewarded!

Thank you for the tip with the Log Viewer. I discovered there an interesting error:
SLD start failure, profile parameter ObjectManager.BufferInstances must be set to FALSE in J2EE cluster environment. Change the SLD profile and start the application again.
Is it possible that the SLD is down and causes my error?
Because in the Adapter-Monitor i get an similar error for the JPR-Adapter (what for is this Adapter?):
SLD access SLD host:port = xxyyzz:12345
Error getting JPR configuration from SLD. Exception: entityNotFound
No access to get JPR configuration

Configuring Maintenance Optimizer issue

Hi All,
This is the first time I am configuring SM 4.0. I am done with installation, and want to conigure maintenance optimizer.
I followed the IMG guide step by step.
But after all done, there are no systems listed under the maintenance optimizer of my Solution Landscape.
Instead I get a message:
Change requests cannot be displayed or created, because the RFC connection has errors, or is not maintained.
Check whether the RFC connection to the Change Man agement system is under the key CHARM_DEST (APPLI field in the DESTINAT field), in the view BCOS_CUST. You need authorization for the cross-client view/table maintenance (S_TABU_DIS and S_TABU_CLI).
Check the RFC connection in the transaction SM59. If your SAP Solution Manager system is also your Change Manager system, the internal connection must be 'NONE'. You need authorization for the administration of RFC connections (S_RFC_ADM).
Meanwhile I have issue with my J2EE engine not starting due to some UME issues, which I am working on parallely.
Please do let me know what would be the issue.
I have followed the note 990534 and doen with the BC sets.
But I am not sure where am I making a mistake.
One more doubt is, does maintenance optimizer connect to the SAP service market place through SAPOSS??? and am I not seeing anu system because my SLD is down. Please bear with my ignorance, I am a little lost here.
I have completed the solution landscape, logical component, etc.
Please help...
Regards,
Sapfan

Hi JP,
My current SP levels are:
SAP_BASIS - 0011
SAP_ABAP - 0011
PI_BASIS 2005_11 - 0011
ST-PI 2005_11 - 0003
SAP_BW - 0011
SAP_AP - 0007
BBPCRM 5 - 0007
CPRXRPM - 0007
ST400 - 0010
ST-ICO - 0008
ST-SER - 0007
The issue is I did not connect to SAP yet, as I have only one RFC generated i.e. SAPOSS> Since my SAPRouter is not yet configured, I will connect to a valid SAPRouter once all the SM config is done.
I als got sone error in generating RFC for the main system when I used the wizard. I selected Change management and customizing directory and also selected Assign RFC for system monitoring.
I changed back the UME source to client 001, but still same.
Moreover the maintenance optimizer is not showing the satellite system.
I am really stuck.
Please let me know if you require more information.
I created the main system and defined the solution, as per the SPRO->Solution Manager -> Advanced Configuration -> Basic settings ->
I started this only after applying SP10.
Now I don see the systems both for Setup Solution monitoring and in maintenance optimizer.
WIll the system not reflect is the java ebngine is down???
Regards,
P.K

Problem SLD down

Hi,
How can I turn off the SLD to update and not to crash the Java applications linked to it?
Regards.

Hi, Diego,
If your Solution Manager is also your Production SLD, then you cannot do what you wish. When the Java stack goes down during the upgrade, SLD will go down also.
SAP does not (any longer) recommend using solution manager as your productive SLD. They recommend having multiple SLDs that all talk to your solution manager SLD.
To work around your problem:
1) if you have another Java AS you could activate, configure and replicate the SLD data from Solman SLD to that SLD, and then point your production systems at it prior to upgrading.
2) if not, then the same can be accomplished by installing a new Netweaver Java AS, and configuring it as your production SLD, and then tying it to solman SLD prior to your upgrade of Solman.
The SLD guide in SDN provides this information and recommendations. - System Landscape Directory (SLD)
Cheers
Phil

CRM 5.0: EP, UME, SLD etc. Down

HI Experts:
This question is a tricky one. I recently finished updating the SPs to the latest level. Everything went through fine. The system also has the latest kernel as well.
Now, when i restarted the system up, everything started up fine. Everything is green....J2EE process table is also green. MMC says that the system is running. However, everything accessed from the SAP GUI works fine. However, I can't access EP 7.0, SLD, UME etc etc.
I get the following error message:
"You have requested an application that is currently stopped"
That makes me want to think that even though it is saying that the Server is running, not all of the processess are running in it. Usually it took around 8 minutes to load up, but now its taking around 2 minutes.
Has anyone encountered/resolved this in the past?
This all started to happen after i updated one of the JAVA patches....so, i feel as though, that must have done something.
I have already reloaded the server and the instances....however, everything is still the same.
Can anyone help me out with this.

Hi Imran,
Some times not All the J2EE Services will be up. If you restarted the Server. So, You need to go Services and right click on "J2EE Service" and then restart all services. It will solve your problem.
Message was edited by: MOHAMMED ABDUL HADI

XI if SLD is down

Similar Messages

Maybe you are looking for