Hi running 1.2.4

On a Dell Server with a Quad Core Xeon X3323 running at 2.5 Ghz, with 16 GB of memory. Running tools like top shows that I am using no virtual memory, I have a 500gb RAID 1 array, and it's only using about 50GB. So tons of space.

Groupwise Server
8.0.2 HP2

(looking at upgrading to Groupwise 2012 in a couple of months)

We have around 120 users on this mobility server, and the vast majority 85-90% are iPhones and iPads. We are having a lot of ciients complaining about attachments not being received, they are getting lots of very small .txt attachments saying they attachments have failed to download.

I had a look at a TID 7009702 that said about them being delayed, so increased numWorkers to 8 in the relevent connector. This seemed to improve things for a couple of weeks then it just started happening again. I tried stopping and restarting datasync yesterday to see if that would improve it and the whole server went mad and stopped downloading any emails at all. I ran through the miantenance routines as per 7009453 and this didn't help. In the end after running out of ideas I literally just rerun the 1.2.4 upgrade script rebooted and that solved it! So i'm back working again. But see quite a lot of errors still in the mobility log, like below:-

2013-03-07 11:10:17.021 ERROR_VERBOSE [ATT_POOL_Thread-14] [gwconnectorlogic:1454] [userID:***] [eventID:] [objectID:] [AttachmentsDownload] Failure sending attachment BLAHBLAH2013_He#2559.tif to engine
2013-03-07 11:10:19.240 ERROR_VERBOSE [ATT_POOL_Thread-16] [gwconnectorlogic:1454] [userID:***] [eventID:] [objectID:] [AttachmentsDownload] Failure sending attachment BLAHBLAH2013_He#255A.tif to engine
2013-03-07 11:11:20.689 ERROR_VERBOSE [ATT_POOL_Thread-16] [gwconnectorlogic:1454] [userID:***] [eventID:] [objectID:] [AttachmentsDownload] Failure sending attachment BLAHBLAH2013_He#255A.tif to engine

Is it worth upping numworkers more... or would that not make the slightest difference, is it a case of upgrading to Groupwise 2012 might make a difference and I should just leave it plodding along until then? Or do you have a receommended maintenance or reboot routine in place to keep yours working fine? Any ideas / suggestions would be appreciated. Is it worth setting up a script each sunday to shutdown rcdatasync and then give the server a reboot? or just reboot it and let it sort out shutting down the rcdatasync itself? or is that just a waste of time.