The BOM doesn't specify 32 or 64, but with most of Cisco CC software always go with 32 bit. Per the BOM 12 gigs is correct.
Oh boy, do I know about this one.
On the CUIS Archiver ... the BOM certainly does not say you can't use x64, and you would think that with a lot of memory (we use 16GB) that would be the way to go.
So you put on Windows 2003 Enterprise x64 and SQL Enterprise x64 and the install goes fine until there is some code in the installer to assign Local Priveleges and it crashes.
So, no you can't use x64.
As David says, there is another BOM, the BOM for Cisco ICM software which specifies only 32 bit. It doesn't mention UIS.
You would not think it would be that hard for Cisco to add this to the UIS document, would you?
Now getting back to x64 and the Archiver.
The Archiver is pure SQL code and it would work perfectly - it would be faster and fully address the memory, rather than the dodgy /PAE which you have to apply in the boot.ini.
But some brain-dead programmer wrote the installer using only 32-bit privelege changing function, so the installer crashes. It's so easily fixed.
Welcome to CUIS. I bet after you have started the install you come back with more questions. ;-) The CUIC part is easy, but the Archiver a little trickier.
Is your ICM 7.5 or 7.2(7)?
One of the real weaknesses in CUIS is the documentation.
Here is an example. When building the Archiver, you need to connect to the HDS so that a scheduled job can pull the historical information.
"Initial Load is the process by which the Archiver populates its databases with Unified ICM datafrom the HDS for the first time. It takes place after Archiver Configuration is complete. Initial Load is "expensive" for the HDS and needs to take place at a time that will not disruptnormal operations."
As you know, there are two databases on an HDS - xxx_awdb and xxx_hds. The Archiver Guide continually refers to getting data from the HDS.
Which DB should the Archiver connect to?
Secondly - if you are using the Archiver to link to a ICM 7.2 system, you would think the installer would take care of the schema differences.
But no - you have to man-handle this yourself (see page 22) through a meta data table. This should be automatic, since the Installer asks you if you are pointing to a 7.2 system. Not acceptable.
Make sure the time of the Archiver is synchronized to the domain and is in the same timezone as the AW/HDS (the whole Central Controller). This is mentioned in the Guide - but it's a real problem if you miss getting this correct. Should be highlighted.
Remember guys, this is a $100,000 software, but when you read the documentation and try to really create some reports, there is just not enough meat there. Not enough examples.
Finally, although a number of TAC engineers have loaded the CUIS all-in-one and know the basics, the number that really understand the Archiver are few.
I install IC Server and Web server successfully. I try to install the archiver on a simple deployement model.
When the setup check the database connection between archiver and hds is failed. We used local syetm account define on both Archiver and HDS server
Did you have any info for that
I finnaly successfully install CUIS. The problem that I have right now is with the first load. I enable the job in SQL and it fail to run. Error in step Load Archiver
Error from SQL
Login failed for user '(null)'
Did you have any anser for that one
Now you are getting into the part of the Archiver that is a little more complex and where the documentation is a little weak.
In you case it sounds like there is an error with the linked server or the account that the Archiver is trying to use to connect to your AW/HDS and make the extraction for the initial load. Could be a simple password error.
What you can do is to run the initial scheduled load by hand. This is described on page 30 etc. of "Cisco Unified Intelligence Suite Archiver User and Schema Guide" 7.5(2) February 2009. But you do need to read between the lines a bit. It would be great if Cisco expanded on the SQL debugging method because there is stuff missing.
it was a password error. I ssuccessuful ran the initial load process from SQL by running the job manualy. I try it once it works. The first time I changed the EarliestDate as mention in the doc. I did it for 3 days just to make sure that all works fine. Since it works a decide to changed the date to get more data(ine that case one month). So I change the EarliestDate with the new one. I manualy run the job again and I did not get any value older then the first time I run it. I only receive the new data from the last time a run the job.
First time run the job was and Friday july 23 and EarliestDate = july 21
Job run successfully.
Second time run the job was today
and EarliestDate = july 1
No data older then the july 21 both new data between july 23 and july 27 was import.
Did you know ist there a way to run the initial load and go earlier in the past