In TC 9.1 gibt es wiederum einen Bug, der es nötig macht "clearlocks" wesentlich öfter auszuführen (Also so schlimm kann "clearlocks" dann doch nicht sein ;-)
Ich lasss Clearlocks immer früh vor Arbeitsbeginn und in der Mittagspause laufen. Bis jetzt keine Probleme. Auch nicht bei einem reconnect nach der Mittagspause.:
R #:
1920889
Date:
Dec-13-2012
CURRENT WORKAROUND
==================
Run "clearlocks -verbose" on a regular basis, or at least at times of known drops in the server Manager PROCESS_TARGET value.
ADDITIONAL DATA PROVIDED: (Y/N)
===============================
Uploaded to PR_DATA is the file IR1920889.zip which contains the file IR1920889.pdf.
The document IR1920889.pdf includes the customers uploaded screen grabs and also describes in more detail, how to reproduce the issue reported by the customer.
ADDITIONAL INFORMATION:
=======================
The customer has experienced situations with their "ready for production" test system, whereby many thousands of these dead processes can occur over a period of days, due to the configured PROCESS_TARGET variations and the feeling is that when it gets to this stage it can also effect other Teamcenter activities.
The anticipated user count for this customer is in excess of 10K and hence the "ready for production" testing has highlighted this scenario due to the large variations of PROCESS_TARGET being tested. The anticipation is that this effect will diminish somewhat when tcserver processes get assigned to users during production usage, but the current behaviour is still considered incorrect.
Solution
Siemens PLM believes that the problem you reported, where warm tcserver processes that were killed due to a dynamic Server Manager PROCESS_TARGET decrement, were not clearing Teamcenter database connections, will be resolved by code changes made to the upcoming release of our Teamcenter software release 10.1.0
Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP