TM1 Forum

qml wrote: I would rather add a bit of code to do just that and prevent the crash from happening

Thank you for answering. You are right; if you developping new TIs ... but if you have an existing productive TM1 system I would rather wait for an interims fix pack than install FP6. Because you have to check/adapt all TIs which are using DatasourceCubeview as a source.
Olaf

Thank you Olaf for bringing this to our attention.
We just experienced this for ourselves in one of our projects and informed IBM about it. Let's see how long they need to issue an interim fix.

We made a huge mistake and upgraded to FP6 and now we are dealing with this bug. The server crashed 3 times this afternoon and after troubleshooting i found out it was relating to a TI Process that users an "empty" view as datasource for which the Skip COnsolidation option was disabled.

Now we are having hard time restoring to the previous version

Olaf Franke wrote:I found a workaround for Turbo Integrator (TI) processes:
- use the new function DisableMTQViewConstruct() in the Prolog if you use a view as a datasource:

You can also disable the MTQ feature for all TI datasources globally by setting MTQQuery=F in your tm1s.cfg.

The option doesn't appear to be documented, but what else is new.

BrianL wrote:

Olaf Franke wrote:I found a workaround for Turbo Integrator (TI) processes:
- use the new function DisableMTQViewConstruct() in the Prolog if you use a view as a datasource:

You can also disable the MTQ feature for all TI datasources globally by setting MTQQuery=F in your tm1s.cfg.

The option doesn't appear to be documented, but what else is new.

We were not able to undo the upgrade to FP6 ( we upgraded FP4 to FP6 2 weeks ago or so ), we tried every single way, IBM had provided us some steps but they did not work. Our problem was with Websheets having Action Buttons. After going from FP6 to FP4, the buttons do not work any more. So frustrating..

So we have to stay on FP6, but the good news is that this new parameter to turn off MTQ for CubeView Data SOurces seems to help us with the crashing issue, so fingers crossed now. But who knows what other nasty bugs are in this crappy FP6

Another update regarding this. The Parameter MTQQuery=F, avoids the crash when the datasource view returns no data, but unfortunately now it created another bigger problem. If the data Source View contains consolidated cells, then when the parameter MTQQuery=F, it considers the view as an empty view ( does not go to Metadata and data Tab )

So we tried the same code ( code that has been working for more than a year and never changed ), when the parameter MTQQuery=F, it does not process the data returned by view, when the parameter is commented out, then in processes the View but it crashes if the view is empty.

We create the views within TI process ( create in Prolog and Delete in Epilog ) so in commented out the deletion of the view to check if the view returns data, and actually the view has all the data it should, but for some reason, the TI process thinks it's empty

Here are the Options of the view

# Include Consolidated Elements
ViewExtractSkipCalcsSet ( cubName , sViewSubsetName , 0 );

#Here we Skip Rule Calculated Values
ViewExtractSkipRuleValuesSet ( cubName , sViewSubsetName , 1 ) ;

#Here we suppress zeros from source view
ViewExtractSkipZeroesSet ( cubName , sViewSubsetName , 1) ;

DatasourceNameForServer = cubName ;
DatasourceCubeview = sViewSubsetName ;

Dear all,

I got a reply from IBM yesterday regarding on this bug "TI crashes when using cube view as data source and MTQ is enabled.

--------------------------------------------------------------------------------------------------------------------------------------------
We have the Interim Fix on top of TM1 10.2.2 FP6 for the scenario described in the following link :

- http://www.tm1forum.com/viewtopic.php?f ... 4&start=20

So if you want to upgrade to TM1 10.2.2 FP6 then also install IBM TM1 10.2.2.6 IF1002 (IF1)
on top of FP6

In case you don't want to upgrade to TM1 10.2.2 FP6 then I will suggest to wait till FP7 is released. But as of now we do not have the release date of TM1 10.2.2 FP7
--------------------------------------------------------------------------------------------------------------------------------------------

But I not able to find the download link for FP6 IF1. Can anyone verify on that?

You usually need to raise an extra service ticket to get access to the fix, if the IF doesn't appear in public Cognos TM1 10.2.2 Fix Lists.

Gabor wrote:You usually need to raise an extra service ticket to get access to the fix, if the IF doesn't appear in public Cognos TM1 10.2.2 Fix Lists.

This is correct. Any access to the Interim Fixes comes through a request. There needs to be a WebID added to the ACLs.

So, when you make this request, you need to tell the IBMer

• Person downloading the Fix
  WebId of that person/entity
  The Interim Fix that you want

Gabor wrote:You usually need to raise an extra service ticket to get access to the fix, if the IF doesn't appear in public Cognos TM1 10.2.2 Fix Lists.

Maybe. But this is a HUGE flaw in FP6. Any customer on or considering FP6 must have it for FP6 to be viable. Really this IF should be on Fix Central.

ardi wrote:Another update regarding this. The Parameter MTQQuery=F, avoids the crash when the datasource view returns no data, but unfortunately now it created another bigger problem. If the data Source View contains consolidated cells, then when the parameter MTQQuery=F, it considers the view as an empty view ( does not go to Metadata and data Tab )

So we tried the same code ( code that has been working for more than a year and never changed ), when the parameter MTQQuery=F, it does not process the data returned by view, when the parameter is commented out, then in processes the View but it crashes if the view is empty.

We create the views within TI process ( create in Prolog and Delete in Epilog ) so in commented out the deletion of the view to check if the view returns data, and actually the view has all the data it should, but for some reason, the TI process thinks it's empty

Here are the Options of the view

# Include Consolidated Elements
ViewExtractSkipCalcsSet ( cubName , sViewSubsetName , 0 );

#Here we Skip Rule Calculated Values
ViewExtractSkipRuleValuesSet ( cubName , sViewSubsetName , 1 ) ;

#Here we suppress zeros from source view
ViewExtractSkipZeroesSet ( cubName , sViewSubsetName , 1) ;

DatasourceNameForServer = cubName ;
DatasourceCubeview = sViewSubsetName ;

I have a client who just upgraded to FP6 and I have noticed the same bug with a slightly different twist. We can export zero-suppressed views with consolidated cells via a TI process but if you try to export the view by right-clicking on it and choose Export then you got one line of data and then the process just stops and says it was successful.

lotsaram wrote:

Gabor wrote:You usually need to raise an extra service ticket to get access to the fix, if the IF doesn't appear in public Cognos TM1 10.2.2 Fix Lists.

Maybe. But this is a HUGE flaw in FP6. Any customer on or considering FP6 must have it for FP6 to be viable. Really this IF should be on Fix Central.

Please find the FP6 IF1 download link below from Fix Central, which I obtained from my customer support. I still not able to access the page though due to access rights issue. Just share it here first while fixing my access issue.

http://www.ibm.com/support/fixcentral/s ... login=true

Do you have further information about the IBM Interim Fix - e.g. the IBM APAR number or the PMR number?

Paul Coggan wrote:Do you have further information about the IBM Interim Fix - e.g. the IBM APAR number or the PMR number?

If I am not mistaken, this is number: PMR 25429,000,778

I think we have multiple PMRs open on this issue for different customers. We also tried the link and it doesn't work. We have added a query directly referencing the build number and link to a PMR to see if we can get it. Really I can't understand why the support team isn't better coordinated. You would think for an issue affecting everyone on the version with multiple PMRs then everyone with the same issue would be notified once there is a build that addresses the issue? But yet the only notification that there is a fix comes from yingchai. (thanks BTW).

lotsaram wrote:I think we have multiple PMRs open on this issue for different customers. We also tried the link and it doesn't work. We have added a query directly referencing the build number and link to a PMR to see if we can get it. Really I can't understand why the support team isn't better coordinated. You would think for an issue affecting everyone on the version with multiple PMRs then everyone with the same issue would be notified once there is a build that addresses the issue? But yet the only notification that there is a fix comes from yingchai. (thanks BTW).

Is this a cumulative install? Which means I need to install the FP6 (up_plananalytics_winx64h_10.2.5260.167_ml.tar.gz) first, followed by FP6 IF1 (up_plananalytics_winx64h_10.2.5260.1002_ml.tar.gz) ?

yingchai wrote:Is this a cumulative install? Which means I need to install the FP6 (up_plananalytics_winx64h_10.2.5260.167_ml.tar.gz) first, followed by FP6 IF1 (up_plananalytics_winx64h_10.2.5260.1002_ml.tar.gz) ?

All Interim Fixes for TM1 require the base FP to be in place. So you will need to install FP6 first and then apply the IF on top.

thanks for the link. Agree with your comments re IBM support co-ordination.

lotsaram wrote:Really I can't understand why the support team isn't better coordinated. You would think for an issue affecting everyone on the version with multiple PMRs then everyone with the same issue would be notified once there is a build that addresses the issue?.

It is not because Support is not a coordinated body. Support is as much in the same boat as the people on this forum. The releases are governed by a multitude of things and the dates are the same. Support only knows when an Interim Fix or Fix pack is due out when we get told.

Bear in mind there are people on this forum covering all of the time zones and Support is the same. There is a Euro Zone, an Americas zone, an Asian zone and an Australasian zone. So, the dissemination of information has a long way to travel and that is after the news has to be generated and checked.

The people that work on these releases do put a lot of effort in to getting them out as soon as is possible, but when you have a load of cats and you want to push them up a hill, the cats have other ideas. The same with code and computers in general.

Whilst your expectations on what should happen might be one thing, the reality of the situation could be something else.

At any one time a Support Analyst might have 50 cases that he or she is dealing with, depending on what product they support. Those 50 cases might be a range of issues with a range of customers with a range of ideas of what should happen next.
This analyst might be in the UK .. He might have 10 other colleagues with a similar number of issues.
Then he might have French & German colleagues, each with their own situation.
They all might have US colleagues, Indian colleagues and Aussie colleagues.

None of them have visibility of the others PMRs and each of the analysts will specialise in an area of support that interests them and they give help for.
So, an analyst in the UK might specialise in performance issue and he/she is siloed in the UK. If he gets a CAFE PMR he might have no idea of the problem solution. But the guy sat next to him does. So they communicate. Collaboration is continuous in Support.

So, coordinating 100 analysts in up to 15 different time zones is not going to happen. What does happen is that when the central development team releases a fix they will disseminate that information to the whole range of analysts.
All the analysts want to do is get their PMRs closed.

There are numerous methods of finding out when Fix Packs and Interim Fixes are due out, and not all of them will come through support, so to lump all of Support together, calling them uncoordinated, is unfair.

So as The Board knows,
The Fix Pack 6 (Interim Fix 2) is out now and it fixes (at least) 6 issues, maybe even more.
I would suggest contacting your IBM Business Partner, your sales team or the Support Analyst that you raised your PMR with for details on how to get it.
You will need

• The name of the person downloading the IF
  The Web ID of that person/entity
  The actual Interim Fix that you are requesting.
  The PMR that it is related to

is there a separate IF2 download for FP6? I am only seeing the full FP6 download. Can someone please post a link?

thank you in advance