Are local workspace corruptions a common thing?

[colmacpro]

We are in the final stretch to migrating from Ilink 3.4 to Winchill/Intralink 9.1 M050. We have been trying to push and pull and go through as many usage scenarios as possible.

Yesterday Pro/E crashed while I was performing a workspace action. I restarted Pro/E and connected to Wintralink. Pro/E would prematurely exit when refreshing the workspace table. I could access other workspaces that I had been working with but not the one that was active when Pro/E crashed.

I had to remove the local cache and re-register the server to get access to the workspace. Any local modifications were lost.

Since this was my first one, would this be considered a local workspace corruption?

If local workspace corruptions are common, What is more likely to cause them?

Joe

WF4.0 M150

Windchill Intralink 9.1 M050

[SteveVinyard]

Hi Joe, I don't see many corruptions with clients on the latest software. When I do see things like this it has a lot to do with the client machine. There are several factors that come into play:


1- Java Version (supported or not? The latest supported version is 1_6.18) - common

2- Bad Java install

a. Java can be a strange beast and not install properly, especially if there were old versions of it laying around

b. Cleanup your "path" environment variable so it doesn't reference old java version, uninstall all versions, reboot, install supported version (the latest one has some new Java security that can be a headache...I'm not sure if it's fully supported (1_6.20)

c. Once installed properly you should be able to go to Start>Control Panel>Java and see the right version

3- Browser Version (most are supported ie 7&8 and Mozilla)

4- Browser security - common though doesn't usually cause crashes

a. Firewalls

b. IE 8 Security features

c. Blocking of running scripts

d. Popup blocker

5- Spyware - common

a. This can cause crashes and just overall weird stuff as we all know.

6- Faulty hardware

a. This is more common that I thought it would be...maybe I just have a few guys that got a crappy video card

b. Faulty network hardware

7- Network settings - depends on if IT has messed with anything. I don't see this very often but it certainly can have an effect

8- Unsupported Pro/E version in relation to Windchill version

a. Very common and can EASILY cause the issues you are seeing

9- Old .wf cache still existing

a. Clean out the old ones located in C:\documents and settings\%username%\local settings\ptc\wildfire...(the path might not be exact but you'll find it...Local settings is a hidden folder btw)

b. I usually set the PTC_WF_ROOT environment variable to a normal location (C:\ptc\working\.wf) to mitigate the issue; however you still need to remove the old .wf still)

10- Server timeouts on large operations

a. This used to cause pro/e to crash, usually people off-site using the VPN on a slow connection. Apache would timeout and Pro/e would drop the connection or something similar and at times crash.

PTC has some methods of creating log files for Pro/e to figure out what is causing the crash. If you open a call they should be able to help.


Steve Vinyard
Application Engineer

[cid:image001.gif@01CB24E1.70E49060]

[jnanesh.sajjan]

Hi Joe,

Yes, this is the same kind of issues I am having since many months. As you said, pro-crash, I can not access (nor even open) the workspace. Generally, when I launch pro-e and when it asks to login to pdmlink, I select "cancel"...which means NOT logging in. Then I go to server registry and then RMB click on other w/s and choose activate....so then log in. This is the only way that I can simply delete the corrupted w/s. Pretty painful process!!


- Sending via mobile. Please pardon the brevity and possible typos -

Thanks & Regards,

Jnanesha KS

T 91-80-4119 0900 Ext: 233
F 91-80-41190901
M 994 500 4179
Quality Engineering and Software Technologies
No 55 l QuEST Towers l Mahadevpura l Whitefield Main Road
Bangalore l 560048 l www.quest-global.com

[jfrankovich]

I have been experimenting with a script that creates a backup each day of the week (Mon, Tues, Wed, etc) of the .wf files (excluding all of the ProE .wfcont content files). That way, if a user has a workspace corruption, at the worst, we can roll back to the last day of the week that the workspaces were ok.
We also updated very recently from F000 (lots of corruptions) to M050 (none so far).

John Frankovich
The GSI Group LLC
www.gsiag.com<">http://www.gsiag.com>

[avillanueva]

How many items were in the workspace at the time of the crash? While I
have seen case were the workspace never recovers, I have seen the
following. For large workspaces (also those with large family tables),
Wildfire contacts Windchill for cache information, essentially a sync
request. This can be a huge request to the database. In the mean time,
Wildfire is waiting for the response, sometimes for minutes and the UI
is locked (not even refreshing the screen). Eventually, a huge download
occurs and Wildfire comes back to life. No matter how many times you
kill xtop and restart, you hit the same event. This may account for
some of your issues but its just its behavior. Yes, better in later
builds.