At the moment, users are beating down our doors complaining about extreme GroupWise client slowness (2-3 minutes to open the mailbox, 30-60 seconds to access cabinet folders, etc.). This started a few days ago. We are slightly behind in our patching - The GroupWise (POA) Agent Build Version is 8.0.2-92377. Further details;

Scheduled GWcheck Maintenance runs are now taking 8-9 hours.

User = All
Action = Analyze/Fix
Files = Message, User, Document
Verify = Structure/Contents
Fixup num = All
Statistics = No
Verbose = No
Upd DiskSpace= No

Dozens of entries in the logs appear like those below;

Blob Checks;
- Error checking file 3e5d1019.000, STORE_FILE_NOT_FOUND (0xC05D)
Final Stats;
************************************************** *******************
Uncorrectable conditions encountered:
---- -------------------------------------------------- -----
49 Invalid QuickFinder index files (deleted).......... 2
Correctable conditions encountered:
---- -------------------------------------------------- -----
39 Unrecognized or invalid files in mail directories.. 13
65 QuickFinder temp files in index directories........ 1
66 GWCheck log files in log directory................. 2
67 Outdated execution records.(notify/alarm).......... 44
77 Attachment records opened but not readable......... 1
93 Unused blob files (deleted)........................ 126
************************************************** *******************
Overall processing time for the entire job: 8:55:31

The current POA log (and old ones) have dozens of similar entries to below;
10:10:58 432 Notifying client at: Connection socket 207 Bag ID 30456
10:10:58 328 Possibly damaged blob in database = user.db

The 8.0.2 HP 3 readme does contain notes about fixes to the connection socket/bag, POA slowness issues, so our thoughts were to upgrade tonight and run a manual GWcheck - is this the best course of attack? Which options should be selected for GWcheck?