FAST full text indexing with Windchill 9.1

[MikeLockwood]

We currently have FAST InStream full text indexing in place and reasonably working for 9.0. So far we have not yet gotten FAST ESP to work with Windchill 9.1 - and have heard of others have problems getting it working. Lots of strange error messages so far and the process just keeps crashing.

We're now trying to decide whether or not to delay our upgrade to Windchill 9.1 (M040)from 9.0 this coming weekend - hoping to get FASTworkingwith 9.1 and so nottaking away from users a functionality that they currently use. But we won't delay it if it appears that FAST is a huge development.

Question: Does anyone have FAST ESPsuccessfully working with Windchill 9.1?

[MikeLockwood]

Just FYI to all, we have delayed our upgrade - tentatively 2 weeks. Working closely w/tech support now. They have identified specific issues with the FAST server processes. Seems that FAST ESP has not in fact been tested with a real data set until now. We'll post an update if we get this working - if anyone is interested.

[cforeman]

Hi Mike,

I have to say that I am VERY unimpressed with FAST ESP. Like you, we ran on InStream with 9.0 with nary an issue. Since going to 9.1 & FAST ESP, we have had to add more memory to the server, upgrade to Win2003 Enterprise edition (to support the additional RAM), and restart the service each night to prevent it from giving up the ghost every couple of days. It's slower, a memory hog, and very unstable.

Nice upgrade, PTC!

-Charlie

[achukwuka]

Charles,

Did you configure your FAST ESP to run on the same server as your windchill apps?

Thanks


Alexius C. Chukwuka
IT Analyst, PDP Systems
John Deere Power Systems
Product Engineering Center
*Voice: 319-292-8575
*Mobile: 319-429-5336
*FaxFax:319-292-6282
*E-Mail: -

CONFIDENTIALITY. This electronic mail and any files transmitted with it may contain information proprietary to Deere & Company, or one of its subsidiaries or affiliates, and are intended solely for the use of the individual or entity to whom they are addressed, shall be maintained in confidence and not disclosed to third parties without the written consent of the sender. If you are not the intended recipient or the person responsible for delivering the electronic mail to the intended recipient, be advised that you have received this electronic mail in error and that any use, dissemination, forwarding, printing, or copying of this electronic mail is strictly prohibited. If you have received this electronic mail in error, please immediately notify the sender by return mail.

[NATHANBROWN]

We are on Windchill 9.1 M20 and have FAST ESP installed. FAST ESP runs on a separate workstation that also is our publisher. FAST ESP takes 2-2.5GB of our 3GB of ram to run and we have had issues when Pro/E tries to publish and crashes. We think FAST ESP and the publisher will need to be run on seperate machines to avoid issues.

[JohannesSchramm]

Hello,

just wanted toshare our experiences with the FAST-Engineshipped with 9.1 M030.

We upgraded from Windchill 8.0 M040 to 9.1 M030 in November 2009 and have trouble with FAST ESP ever since.

The FAST-ESP-service is not crashing completely. Searching is running ok, but almost every day the indexingdispatcher process crashes after it hascompleted about 1950 indexing jobs. It seems that the indexingdispatcher is accumulating threads to do its job, but does not release them any more. So after about 1950 jobs the CPU obviously runs out of resources and the process crashes.

This leads to the effect that the IndexingQueue-Jobs on our Backgroundmethodserver cannot be processed any more and the queueentries accumulate.

Our FAST-ESP is running on a separate server with Windows 2003 R2 PAE Enterprise Edition and has 8 GB of RAM.

It is easy to restart the indexingdispatcher process of FAST via the web-interface, and all runs well after this but since we have to do this EVERY DAY ist is a PITA!

PTC Tech support has no clue since we opened a call indecember 2009. But meanwhile they have forwardedthis to the FAST support. I'm still waiting for a clue or a fix to this problem. If we get a solutionI will post it here.

However I would not delay a roll-out of Windchill 9.1 altogether because of this. This issue does not affect the end-user. Only the poor administrators are keptoff from doing more important work!

Good luck!

[pnierkamp]

Hello everybody,


As sad as it is I can only confirm Mr Schramms experiences with the FAST ESP Engine and PDM 9.1

We're running / operating a customer installation with PDM9.1 and FAST ESP on a dedicated machine.


The FAST ESP Service is as far away from stable and reliable as it possibly could be.
Miscellaneous problems occurred at the customers installation (which was originally done by PTC themselves):

- The FAST ESP service itself stops responding/working

- Miscellaneous subprocess stop working (with reporting a failed state and without)

- Miscellaneous Indexing Queue Problems occur

o Failing entries

o Entries not being processed although within state 'ready'

o The queues itself becomes sometimes irresponsive

o BackgroundMethodServer getting irresponsive / failing

Current solution to reduce the number of problems occurring:

- Periodically automatic restart of FAST ESP

- Windchill scheduled jobs, that takes care of queue entries and queues

o If accessing BGMS/BGMS queues fails: BGMS restart