TM1 Forum

nick10598109 wrote:

Alan Kirk wrote:But other than that it ran pretty smoothly. I installed the whole smash, and it took about 26 minutes all up. I had the same experience as Steve with item 167; whatever it is it took the bulk of the time though mine was 167 out of... 288 I think it was. The end dialog reported that the Configuration Tool completed with warnings and errors but there was nothing in the log about what those might have been.

The place where it looks 'stuck' is caused by the large number of documentation files that it is copying over and that it is installing all languages. This is expected behavior for the 10.1 installer.

Wouldn't surprise me. This particular "interesting" piece of design has been mentioned, scathingly, by various people elsewhere. In the overwhelming majority of cases the application will run in a single language. In the next largest it might be two languages. But in all cases IBM needs to establish its global street cred by ensuring that I have access to every conceivable language probably even including the Sumerian help files just in case I decide to send a few legions into the fertile crescent to re-establish a branch office. Thus do we have our bandwidth, disk space and time wasted by 2 Gig install packages which clunk along like a prototype steam locomotive (on their way to installing data files into invalid locations like the Program Files path).

IBM: Building Tomorrow's Bloatware, Today.

nick10598109 wrote:What were the warnings and errors in the logs?

Alan Kirk wrote:but there was nothing in the log about what those might have been.

Emphasis added, though there was a typo there; I should have said "logs" since I checked the system logs as well.

nick10598109 wrote:Using Windows Server 2003?

You think that I have far, far more faith in IBM than I actually do. It was a Windows 7 standalone. There is no way on earth I let a new version anywhere near a server until I have a far better idea of what I'm dealing with. If an install blows up a stand-alone it's relatively easy to put it back together again. If it blows up a server, then Hercules' work detail at the Augean stables comes to mind.

nick10598109 wrote:You can ignore these errors.

Waaaay ahead of you, but thanks anyway.

One very basic fix I would like to see is Control C work first time in the TI editor.
Anyone know if this has been fixed?

TM1 Operations Console (ToC)

Right royal PITA to get installed - Ben Hill has done a good writeup and also a comparison to (his) Cubespy:

Operations Console: http://blog.tm1tutorials.com/2012/02/13 ... s-console/
Cubespy: http://blog.tm1tutorials.com/2012/02/14 ... s-cubespy/

And yes, you can use it to monitor pre 10.1 servers and kill runaway processes - the 'Healthcheck' function doesn't work (whatever it does).

One tip - I am testing in a VM with IPV6 active. When entering 'localhost' for the admin host ToC helpfully turned the entry into
0:0:0:0:0:1 or similar, and it couldn't see any servers on that IP. Entering 127.0.0.1 worked better.

ADMIN NOTE: tidied up the links

It's not actually that bad to install, just pedantic and as usual poorly documented. Can't help but read those documents negativly towards the OC, i wonder why... Also fails to point out the OC is free, whereas cubespy is not.

The fact Cubespy has to be installed on a client machine is also a major pain in a company like this. Nothing gets on a PC without being certified, and that costs time & money. For the handful of people that have admin access, this place won't pay for it to be done.

I've been running it looking at 10+ 9.5.2 servers and its been fine. The health check tells you what the server is doing (start up, offline, busy, idle etc) but i didn't have that before so i'm not missing it now. It could do with some work on the GUI side (sizable columns being one current annoyance) but it's a big leap from the command prompt TM1top.

There is also a bug on version control, IBM have passed it on to engineering. If a server config users clientmin/max settings the console doesn't read the values correctly and may or may not allow you to log in to it. Easy solution is to remove those settings.

Alan Kirk wrote: But in all cases IBM needs to establish its global street cred by ensuring that I have access to every conceivable language probably even including the Sumerian help files just in case I decide to send a few legions into the fertile crescent to re-establish a branch office.

LOL! Hilarious.

According to IBM Cuneiform ROCKS

TM1 10.1.0 Performance modeler seems the perfect forward way for all the people who are IBM Cognos EP experts , as of now with TM1 10.1.0 Performance modeler seems to be the replacement tool for Cognos EP Analyst , though not completely.Creating a planning model is way simpler now to build using Performance modeler , i have been testing and trying to work on Performance modeler where it's much easier to build dimensions , cubes and the best of all is the creation of Dlinks (normal , lookup , accumulation etc) all possible which was a quite a difficult manual process writing rules and TI process.Creating a process to transfer data from various sources is much simpler without the need to use TI(Turbo Integrator). They have also introduced quite a couple of BIFs like Cumulate ,decumulate ,lag etc and i am sure there will be many more added with the new versions to be released.Being an EP expert and then switching to TM1, for me personally was sought of nightmare , but with Performance modeler hats off..i am really impressed...Cheers IBM for introducing Performance modeler...
Jacob Joseph

jameswebber wrote:One very basic fix I would like to see is Control C work first time in the TI editor.
Anyone know if this has been fixed?

It actually works and has even worked before, you just have to press the combination twice

Edit: sorry, I understood he meant: 'work for the first time' ...

That's why James' posting used the phrase 'first time' .

Actually there is a good link here
http://www.tm1forum.com/viewtopic.php?f ... 778#p27349
Andy Key says:

An IBMer explained this one to a bunch of us over beer the other week (see Kyro's sig for more details on the beer). You only have to hit it once, but you have to make 100% sure that you release the C before you release the Ctrl. It has to be Ctrl key down, C key down, C key up, Ctrl key up.

I have been doing this since then, and it does work on only one press.

Will try that. Still annoying though.

Jacobsonsoft wrote: TM1 10.1.0 Performance modeler seems the perfect forward way for all the people who are IBM Cognos EP experts , as of now with TM1 10.1.0 Performance modeler seems to be the replacement tool for Cognos EP Analyst , though not completely.

Creating a planning model is way simpler now to build using Performance modeler , i have been testing and trying to work on Performance modeler where it's much easier to build dimensions , cubes and the best of all is the creation of Dlinks (normal , lookup , accumulation etc) all possible which was a quite a difficult manual process writing rules and TI process.

Creating a process to transfer data from various sources is much simpler without the need to use TI(Turbo Integrator). They have also introduced quite a couple of BIFs like Cumulate ,decumulate ,lag etc and i am sure there will be many more added with the new versions to be released.

Being an EP expert and then switching to TM1, for me personally was sought of nightmare , but with Performance modeler hats off..i am really impressed...Cheers IBM for introducing Performance modeler...
Jacob Joseph

I agree - generally. If anything I would say IBM have done their d@mndest to make TM1 into EP, presumably to tempt the EP community to buy back in. But, more or less by the same token, I don't think the TM1 community will find much to attract them. It's really a question of what you know, and (having worked mainly with TM1 and lightly with EP) to me, they are just two routes to the same destination. Certainly the rules generated by the 'Dlinks' are truly dreadful. I've yet to dig into the generated TIs.

Testing out TM1 Applications (the artist formerly known as TM1 Contributor) on a VM under Server2008, I'm finding some weird behaviour:
I'm mainly working with IE9 with the silly secure mode switched off. Using that, the right-click menus are masked by IE's rightclick menu, but with some fiddling you can get to the options of
TM1 Application Web
Cognos Insight Connected
Cognos Insight Distributed
The latter two work. I'm on the server so have no way of telling what's going on with the 'Distributed' option. But selecting TM1 Application Web fails to initialise the input view.
But...
Using Chrome, the rightclick menu isn't covered by cr@p and TM1 Application Web runs fine. However the Insight options ask to download and install CognosInsight.msi and this fails to complete.
Doubtless Chrome isn't a supported environment - and I've read the issues with IE9 on 9.5.x. But I'd certainly not want to show this shiny new tool to a prospect as it stands, so suggestions are very welcome.

Saw the new eye candy today and it is certainly very pretty. The move away from views and subsets to a pure MDX query approach is certainly interesting but it seems to bring a lot of benefits.

The modelling interface was certainly great demo and presales ware, although after a PoC I can see a real need to go back and re-code everything properly. The dependence on CAM security is something that gives me the shudders though.

I imagine you were 'attending' one of the two virtual launch sessions. I was 'at' the first one.
I've been going through 10.1 recently:
a Views and subsets are very much still around indeed in a number of areas you can't work with TM1 data if you haven't defined them.
b I agree 100% with you about CAM's awfulness - but in the sample app GO_New_Stores which came with the install the tm1s.cfg file has no IntegratedSecurityMode so defaults to 1, and works fine with Modeler and Insight. I also dropped my regular demo server on the VM, which uses 1, and this works too - I've been testing out submissions etc using userids I already had.

Browser issue update....
Installed Firefox 10.0.0.2, right-click works properly, all three options (Web/Insight online/Insight offline) work. Micro$oft...
Support matrix says:
Firefox 3.5​, 3.6,4​ Compatible​​
Firefox 5,8 ​(6)​ Active​​

Note (6) - love the weasel words here
Note that product testing will be conducted on a periodic basis, and may or may not align with the Mozilla Firefox release schedule. It is our best estimation that changes in future versions of the Firefox browser will not cause issues with our product used as long as Mozilla does not remove, explicitly or inadvertently disable functionality that the browser based tools rely on. However we recommend that you contact IBM support if you do encounter any such issues.

David Usherwood wrote:Note (6) - love the weasel words here
Note that product testing will be conducted on a periodic basis, and may or may not align with the Mozilla Firefox release schedule. It is our best estimation that changes in future versions of the Firefox browser will not cause issues with our product used as long as Mozilla does not remove, explicitly or inadvertently disable functionality that the browser based tools rely on. However we recommend that you contact IBM support if you do encounter any such issues.

I think this is a reasonable statement on IBM's part. It is Mozilla that have gone for the 'It's Friday, it must be time for a new version of Firefox' release schedule.

I agree - but it was more the turn of phrase which caught my eye

Playing with Insight I have found how to edit 'real' rules rather than the 'lite', line-orientated version - 'Set' cube rules on the first Widget menu. I'm pleased, but puzzled, to see that the line-based rules are not cluttered with the dreadful stuff which comes with the Performance Modeller line-based rules eg: Wonder why?

Perhaps a minor issue, but one the accountants won't be keen on, is the fact that it seems to be impossible to make subset elements expand upwards in performance modeller. Even if it is manually selected in architect and saved as a subset, the rollups break in PM.

Improved documentation of configuration parameters is certainly welcome!
http://www-01.ibm.com/support/docview.w ... RXT&mync=E

I'm sure some are still missing, but it is a good start.
and finally we have StartupChores

Cheers Lotsa. Looks very interesting indeed.