Community Tip - Did you get called away in the middle of writing a post? Don't worry you can find your unfinished post later in the Drafts section of your profile page. X
Hello Everyone,
We went live with Windchill 9.1 M060 last week after a migration from Pro/INTRALINK 3.4. We are experiencing extremely slow check in times for some drawings and assemblies. For example, one particular drawing and its assembly (two objects checked in) will take 45 minutes for the check in to complete. The upload of the objects is fast (~30 seconds) but the check in portion of the process seems to hang. This is only occurring for certain large assemblies and drawings. I don't see anything in any of the log files to indicate a problem. There is no spike in resource or memory usage on the servers when there is a hanging check in.
Has anyone else experienced a similar problem? Any thoughts on why the check in portion of this process is taking so long for certain objects.
Thanks,
Steve
First off I would like to thank everyone that took the time to respond to my question. There were many good ideas on what to look for and additional client and server tuning tips. I received a number of replies that stated they had or currently have the same problem as we are seeing.
Our problem was resolved by allocating more memory to Oracle. We bumped the memory allocation for Oracle from 2 GB to 12 GB and the check in time for our test drawing went from 45 minutes down to 20 seconds.
Thanks again to all for the help.
I had a request to post the Oracle changes that were made back to the exploder so please see below.
Our environment - Oracle 11G R2
Please note: An Oracle DBA at my customer made these changes. Posted below are the changes that were made. It is not clear if one change or a combination of the changes below provided the improved performance. I would recommend having an Oracle DBA make these changes to your environment. One reply to this thread provided contact info to a highly recommended Oracle DBA that I can pass along if there is interest.
For our environment, we made the following changes:
wind.__db_cache_size=1828716544 --> 11509170176
wind.__java_pool_size=16777216 --> 33554432
wind.__large_pool_size=16777216 --> 33554432
wind.__pga_aggregate_target=2147483648 --> 6442450944
wind.__sga_target=2147483648 --> 12314476544
wind.__shared_pool_size=268435456 --> 671088640
*.processes=500 --> 1000
*.sessions=225 --> 800
*.sga_max_size=2147483648 --> 12288800000
*.sga_target=2147483648 --> 12288800000
These parameters were added:
*.db_files=400
*.db_writer_processes=8
*.session_cached_cursors=100
*.session_max_open_files=20
*.sort_area_size=8192000
I hope this helps.