03-05-2015 08:08 AM
We are in the process of migrating from stand alone copies of ACT (2009-2013) to ACT v16 sp1 Premium, around 70 users. We have a dedicated server running Windows Standard 2012 r2, SQL 2008 R2 SP3 on a hyper-v. It has 40 GB of RAM dedicated to the HV, RAID 10 for SQL Database and another raid 10 volume for the SQL logs. It has a Dua Xeon CPU with 12 total cores, 8 of which are dedicated to the HV.
The issue we are having is the ACT! server will max out the CPU at %100. It appears it is happening when multiple try to sync. If we restart the Network Sync service, the CPU drops down to 0-5%. After a few more people start syncing, it pushes back up to 100%. By a few people, I mean 3-5 at the most.
I have run the index/repair through ACT/ACTDIAG multiple times. I also have run SQL DBCC and it comes back clean.
Is there any other troubleshooting steps we can take?
03-05-2015 09:29 AM
Are you running SQL 2008 R2 Standard or Express?
03-05-2015 12:28 PM
I've seen that on a lot of systems. I think that when they wrote the sync routine that they didn't put in code to allow othe things to happen on the system at the same time. I've found that if I set the afinity of the process to a smaller set of CPUs and lower the priority it will actually improve performance. Give that a try and let us know how it works for you.
03-06-2015 11:32 AM
The changes made a big difference in CPU utilization. We have not hit above %40 since. Syncs are still slow though -
[ Info | 3/6/2015 11:10:20 AM ]Message: Sync remote client created.
[ Info | 3/6/2015 11:10:20 AM ]Message: Checking if schema changes are available for sync.
[ Info | 3/6/2015 11:10:21 AM ]Message: Client establishing synchronization with server database: IDENTISYS
[ Info | 3/6/2015 11:10:22 AM ]Message: Initial Handshake completed.
[ Info | 3/6/2015 11:10:22 AM ]Message: Checking if data changes are available for sync.
[ Info | 3/6/2015 11:10:24 AM ]Message: Client initialized database send session: 3
[ Info | 3/6/2015 11:10:33 AM ]Message: Client initialized.
[ Info | 3/6/2015 11:10:33 AM ]Message: Client schema synchronized.
[ Info | 3/6/2015 11:11:01 AM ]Message: Server initialized.
[ Info | 3/6/2015 11:11:12 AM ]Message: Server schema synchronized.
[ Info | 3/6/2015 11:15:11 AM ]Message: Files synchronized.
[ Info | 3/6/2015 11:28:54 AM ]Message: Content synchronized.
[ Info | 3/6/2015 11:29:02 AM ]Message: Exceptions synchronized.
[ Info | 3/6/2015 11:29:04 AM ]Message: Synchronization completion done.
[ Info | 3/6/2015 11:29:06 AM ]Message: Synchronization has completed.
[ Info | 3/6/2015 11:29:06 AM ]Message: Items sent:
Activity - Updated 36
Security - Added 1, Updated 2
Total Records Synchronized - 39
Files - Attempted 8, Completed 8
Exceptions Occurred - 0
[ Info | 3/6/2015 11:29:06 AM ]Message: Items received:
Activity - Added 29, Deleted 18, Updated 234
Company - Added 1, Updated 3422
Contact - Added 2, Updated 75
Group - Updated 14
History - Added 372, Updated 5909
Metadata - Added 1, Updated 4
Opportunity - Added 42, Updated 2089
Security - Added 39, Deleted 11, Updated 84
Sync - Added 252, Updated 107
System - Added 2, Updated 2
Total Records Synchronized - 12709
Files - Attempted 10433, Completed 10433
Exceptions Occurred - 1
03-26-2015 12:20 PM
Update on this - We continued to have slow syncs. I ran a test restore as on our Live DB. It shrunk the DB size from 11.9GB to 4.8GB. Things were working great for about a week. We are now seeing syncs acting strange again. A person can have a new RDB, run a sync and there are 1000's of updates. These syncs finish in 3-4 hours. Then we run another, and a few 1000 more records. Nothing has changed in the Live DB during this time.
I ran another restore as today. The DB shrunk down from 5.9 to 4.9 GB. I am guessing corruption? We run have scheduled a task for maintenace that runs nightly. But never reports any problems.
act 16.1 hf 2.
03-31-2015 01:47 AM
With this number of databases syncing you need to run a check and repair on all databases weekly. Set up ACT scheduler to do this. Make sure that only live database are in the list of databases under Manage Databases. You may need a cleaning tool as well but try this first.
03-31-2015 09:55 AM
To make sure I understand, have each user run a Check/Repair on their RDB each week through the ACT scheduler.
Where would I look for the list of databases under Manage Databases? I am not sure what you mean here.
I would be interested in looking at cleaning tools to get an idea of what they do. What do you recommend?
Thanks for the reply!