I thought the hidden design setting from one server would not replicate to another. Therefore it would be possible to maintain a server (e.g. as Hub server) with the design visible whilst the App servers would have the hidden design.
I’ve deployed a database to server A. Created a replica onto server B. Then replace the design on server B but check the ‘Hide formulas and Lotus Script’.
But after the first replication the replica on Server A (the unhidden design), has a visible design elements but the Lotus Script and Formula in the agents cannot be seen. All other design elements seam ok.
bug? <in small letters. I don’t want to cause an outrage as always seams to happen when hidden design is mentioned in this forum>
Subject: Agent Hidden design and replication.
Hiding the design is not just a setting. It removes source code and encrypts formulas. This change made the agents on B more recent, so these agents with no source code replicated back to A. On A, the database does not have the hidden design setting, so you can still see the design elements there.
Subject: RE: Agent Hidden design and replication.
Thanks for that. You are of course correct and It is pretty obvious really.
I was replacing the design on Server A with the same template as used to hide the design on server B, the same template used for the original deployment. But since there were no changes to the template there were no design elements that needed updating on server A. Hence the most recent changes, those that hid the design on server B, are visible.
Modify the template design elements, simply signing them with a new ID is enough, then replace the design on Server A and ‘bingo’ we have the code back in the un-hidden design.
However, during my testing I noticed that if I open the NSF file (hidden design) with a text editor I can still see the Lotus script code for the agents and Script Libs etc. Is this right if the hide design option removes source code.
TIA
Subject: RE: Agent Hidden design and replication.
The source code should be completely removed, leaving only the compiled code in the NSF file. Parts of the compiled code are readable – for example, string constants – but you should not be able to read the rest of the code. If you can, you might have found a bug…
Subject: RE: Agent Hidden design and replication.
Oh…
I’d like to know if anyone else can see ‘hidden’ code.
I’ve opened an nsf (and ntf) with hidden design in UltraEdit. Scroll down (there is a lot that cannot be read) but eventually you can see LS code in clear text.
Subject: RE: Agent Hidden design and replication.
Rod, furthermore
I am also able to step through LS code in the debugger for a database that has is design hidden. Surely this is not right.
In 4.6 I’m sure that you could not step through code with a Db having its design hidden.