cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Showing results for 
Search instead for 
Did you mean: 

We are happy to announce the new Windchill Customization board! Learn more.

cache corruption practice

Dmi3U
15-Moonstone

cache corruption practice

Hi Pros,

I am wondering if you have a practice in your company about removing/refreshing cache on client and how are you dealing with corruptions.

Thank you,

Dmitry

5 REPLIES 5

Our procedures specify that all files must be checked and workspaces
deleted once a week.
This reduces the chances of corruption coming into a workspace.
We also do file renaming per our Bill Structure and all files must be
removed from workspaces for the rename to work.


Ben Loosli
Sr IS Technologist
Engineering Automation Systems
L-3 Communications/Integrated Systems
Waco, TX 76715

254-867-4069
-


We currently deal with corruptions as they occur. Generally the corruptions have presented themselves by aberrant behavior and the user will report it. I'll have them upload their work and clear the cache, fixing the problem. There have been cases where users have not been able to upload their work due to the corruption, and that work had to be backed up to the hard drive, then loaded back into Windchill after the cache was cleared. There are further cases where the work could not even be opened properly to do the backup, and in those cases the work had to be manually recovered from the cache (there is a tool in the Files section here that does that pretty well). Sometimes the corruption is so deep that it requires a new workspace be used in addition to clearing the cache. So far I've had 4 corruptions reported this month, out of 119 seats of ProE.

James Avis

NASA JSC

mskaggs
9-Granite
(To:Dmi3U)

Calling all PDMLink experts,

This thread is quite timely. Last week one of our users had a major WS corruption.
We are struggling to figure out what was done wrong, if anything.
Here is the scenario:

The user has a large released assembly added to the Workspace and wants to develop a parallel design.
He is not certain which parts of the assembly need to change, so makes
local modifications to the component objects. Once the new design is
somewhat solidified, he attempts to copy some of these objects to new names.
In order to do this, the objects must be checked out and uploaded to the server-side WS.
The check out action succeeds; but the upload action fails with this error:
java.lang.NullPointerException

What's worse the objects are no longer retrievable in Pro/E.
Also, exporting the objects from WS fails.
The WS view for these objects still has a plus sign in the 'Local Workspace Status' column
and a yellow check mark in the 'General Status' column.
The local cache was then cleared, with no change to the WS view.
At this point a support call is opened with PTC, and via Webex session
the PTC engineer can verify the issue. He suggests looking in the local cache folder
for the lid*.wfcont files. There are none.
The chosen recourse is to delete the .wf folder and drive on, thus losing all of the local modifications.

Anyone seen this? Any suggestions other than avoiding local mods?
What is clearing the cache supposed to do other than reduce the size of the files in .wf?

Thanks for listening,

Mike Skaggs

Scriptpro

You said the object were "add to Workspace". Do you mean "checked out to a Workspace"? Herb MillerTWeatherford, Inc.3148 Beach BlvdCicero, IN 46034www.tweatherford.com 866 936-2756 x 305502 552-7179 mbl419 821-5765 faxRead our latest newsletter by clicking here"If you continue to ignore PLM, you will learn about it from your biggest competitor"----------     
mskaggs
9-Granite
(To:Dmi3U)


In Reply to Herb Miller:

You said the object were "add to Workspace". Do you mean "checked out to a Workspace"? Herb MillerTWeatherford, Inc.3148 Beach BlvdCicero, IN 46034www.tweatherford.com 866 936-2756 x 305502 552-7179 mbl419 821-5765 faxRead our latest newsletter by clicking here"If you continue to ignore PLM, you will learn about it from your biggest competitor"
Top Tags