TM1 Web 2.0.6 - Bogus Timeouts

Post Reply
User avatar
Alan Kirk
Site Admin
Posts: 6606
Joined: Sun May 11, 2008 2:30 am
OLAP Product: TM1
Version: PA2.0.9.18 Classic NO PAW!
Excel Version: 2013 and Office 365
Location: Sydney, Australia
Contact:

TM1 Web 2.0.6 - Bogus Timeouts

Post by Alan Kirk »

Is anyone else using TM1 Web in PA 2.0.6?

A new parameter was introduced to the web config file in 2.0.6 to deal with the timeout. If I may quote:
The default TM1® Web session timeout is 20 minutes. When TM1 websheets are deployed to IBM® Planning Analytics Workspace, you might encounter TM1 Web session timeouts. You can modify this setting in your environment.
No, that's not my problem. Though I have been told that I will be getting Windows Server upgrades I do not currently have them. As a result of which I do not use the Docker-based abomination called PAW.
As of IBM Planning Analytics Local version 2.0.6, you must not change the session-timeout value in the web.xml file.
So, IBM, this is you telling me that you have your shiny new parameter working right. Right? RIGHT??
In IBM Planning Analytics Local version 2.0.6, there is a parameter in the tm1web_config.xml file called HttpSessionTimeout. You can use this parameter to customize the session timeout (in minutes) of the HTTP session for TM1 Web.

If the HttpSessionTimeout parameter is not specified (missing or blank), the value is less than 1 or not a numerical value, the default session-timeout that is defined in the web.xml file is used.

If you are using IBM Planning Analytics Local version 2.0.6 or later, to customize the session timeout for TM1 Web, set the HttpSessionTimeout parameter in tm1web_config.xml. See step 1.

If you are using IBM Planning Analytics Local version 2.0.5 or earlier, to change the default session timeout, set the <session-timeout> parameter in web.xml. See step 2.
So riddle me this. The default tm1web_config.xml had an empty string in it. And the timeout was not 20 minutes, it was barely even 20 seconds. In fact it was as soon as I either clicked on an application folder or breathed on the frapping thing that I would get a timeout message.

So I set it to 60 as per the example, even though I hate timeouts on principle. Still, the very first thing you do after you log in... timeout. I changed it to some astronomical number. Timeout within seconds. I set it to 20. Timeout within seconds.

Thankfully I still had one 2.0.5 instance floating around which actually works and as a workaround I pointed the production web servers back to that.

I haven't raised this as a bug with IBM (yet) because I won't be able to reboot the server until the wee smalls of tomorrow morning and also I want to see whether anyone else is encountering this.
"To them, equipment failure is terrifying. To me, it’s 'Tuesday.' "
-----------
Before posting, please check the documentation, the FAQ, the Search function and FOR THE LOVE OF GLUB the Request Guidelines.
lotsaram
MVP
Posts: 3651
Joined: Fri Mar 13, 2009 11:14 am
OLAP Product: TableManager1
Version: PA 2.0.x
Excel Version: Office 365
Location: Switzerland

Re: TM1 Web 2.0.6 - Bogus Timeouts

Post by lotsaram »

Yeah this bug in TM1 Web really get's my goat. (It isn't confined to just 2.0.6 BTW but crops up in all builds since Websphere replaced Tomcat from what I can tell).

A colleague gave me a tip just a few days ago that a workaround to the "instant TM1 Web timeout" is to reset the user's password. Then for some reason it doesn't happen. (I haven't yet tested to see if this actually works).
Please place all requests for help in a public thread. I will not answer PMs requesting assistance.
User avatar
Alan Kirk
Site Admin
Posts: 6606
Joined: Sun May 11, 2008 2:30 am
OLAP Product: TM1
Version: PA2.0.9.18 Classic NO PAW!
Excel Version: 2013 and Office 365
Location: Sydney, Australia
Contact:

Re: TM1 Web 2.0.6 - Bogus Timeouts

Post by Alan Kirk »

lotsaram wrote: Thu Dec 20, 2018 9:27 am Yeah this bug in TM1 Web really get's my goat. (It isn't confined to just 2.0.6 BTW but crops up in all builds since Websphere replaced Tomcat from what I can tell).

A colleague gave me a tip just a few days ago that a workaround to the "instant TM1 Web timeout" is to reset the user's password. Then for some reason it doesn't happen. (I haven't yet tested to see if this actually works).
Innnteresting. Oddly I haven't encountered it at all in 2.0.5 and assumed that it had something to do with the Great! New! Change! I think I'll raise a ticket on it then but to cover my bases I'll make sure that I've said that I've at least tried a reboot, not that I expect it to do a single thing.
"To them, equipment failure is terrifying. To me, it’s 'Tuesday.' "
-----------
Before posting, please check the documentation, the FAQ, the Search function and FOR THE LOVE OF GLUB the Request Guidelines.
TJMurphy
Posts: 74
Joined: Mon May 12, 2008 12:25 pm
OLAP Product: TM1
Version: PA 2.0.6 Local
Excel Version: Excel 2016

Re: TM1 Web 2.0.6 - Bogus Timeouts

Post by TJMurphy »

We just upgraded our system to PA 2.0.6 over the weekend and a couple of users (well, me so far but I'm expecting more as they get to the office today) hit this one. Ops Console was fine by the way.
The password change trick didn't work for me (either via Perspectives or Web). But what does seem to have worked was to log on using IE which was ok and then logging on again via Chrome.

Does this tie with anyone else's experience?
User avatar
Alan Kirk
Site Admin
Posts: 6606
Joined: Sun May 11, 2008 2:30 am
OLAP Product: TM1
Version: PA2.0.9.18 Classic NO PAW!
Excel Version: 2013 and Office 365
Location: Sydney, Australia
Contact:

Re: TM1 Web 2.0.6 - Bogus Timeouts

Post by Alan Kirk »

TJMurphy wrote: Mon Jul 01, 2019 8:21 am We just upgraded our system to PA 2.0.6 over the weekend and a couple of users (well, me so far but I'm expecting more as they get to the office today) hit this one. Ops Console was fine by the way.
The password change trick didn't work for me (either via Perspectives or Web). But what does seem to have worked was to log on using IE which was ok and then logging on again via Chrome.

Does this tie with anyone else's experience?
I should have added a piece of information to this, but by the time I found out I had completely forgotten about having posted this thread. My problem occurred because I use a web-based piece of server monitoring software which I won't name because the blame does not lie there.

Around 2.0.4, apparently, IBM renamed the TM1Web session cookie with the same name as the one that this monitoring software used. This resulted in a conflict where Web would kill the monitoring software and the monitoring software would kill Web. It was dealt with in a subsequent upgrade to said monitoring software.
"To them, equipment failure is terrifying. To me, it’s 'Tuesday.' "
-----------
Before posting, please check the documentation, the FAQ, the Search function and FOR THE LOVE OF GLUB the Request Guidelines.
Post Reply