Initialize Synch Crash after 30-40 minutes on version 627

[Former Member]

I have installed the new version of webtools (627) and the Synch Manager crashes after 30-40 minutes after starting the initial synch process. No error messages are given, it simply unloads. I have tried a number of times following a server reset etc with the same result.

The synch manager log shows the following as the last entry each time:

<event>

<messagetype>I</messagetype>

<date>2008-04-16 12:43:22.71</date>

<synchmodule>Synch</synchmodule>

<message>Starting SBOPartner synch.</message>

<synchmethod>SynchAll()</synchmethod>

<objecttype />

<objectid />

</event>

I have also run SQL Profiler to try to get some insight into what is going on.

The last TSQL caught by the profiler is:

exec sp_executesql N'SELECT LineNum FROM CRD1 WHERE CardCode = @cardcode AND Address = @address AND AdresType = @type',N'@cardcode nvarchar(8),@address nvarchar(32),@type nvarchar(1)',@cardcode=N'C21581D2',@address=N'Area 5- Malahide Dent-DO NOT USE',@type=N'S'

I have looked at this customer and their addresses and cannot see anything unusual.

The process that I followed was:

Uninstall the previous version via control panel - add remove.

Install the new version (Installer and Synch manager)

Using the installer create a new instance (B627)

Restore a new copy of the Promed SAP db (Promed627)

Setup salesperson, customer, and payment term in SAP

Open synch mgr, install plugin into Promed627.

Setup other Synch mgr parameters and table values.

I used this process because I did not want previous versions to cloud any issues. Therefore I started with a fresh install and fresh data.

I have checked the installer and synch dll's and they are all version 627.

[Former Member · Accepted solution]

Ran the synch again with different userid and password generation settings and received the following errors before the system crashed:

<event>

<messagetype>E</messagetype>

<date>2008-04-16 17:51:15.00</date>

<synchmodule>SBOPartner</synchmodule>

<message>Attempted to read or write protected memory. This is often an indication that other memory is corrupt.

at SAPbobsCOM.IBusinessPartners.Update()

at NetPoint.SynchSBO.SBOObjects.SBOPartner.SBOToNetPoint(SBOQueueObject qData)

at NetPoint.SynchSBO.SynchObjectBase.Synch()</message>

<synchmethod>SBOToNetPoint()</synchmethod>

<objecttype>2</objecttype>

<objectid>C20257</objectid>

</event>

SQL Profiler:

xec sp_executesql N'UPDATE PRX_Transaction_Queue SET synchstatus = @status, syncherror = @err WHERE object_type = @otype AND list_of_cols_val_tab_del = @keys ',N'@otype int,@keys nvarchar(6),@createdate datetime,@status nvarchar(6),@err nvarchar(296)',@otype=2,@keys=N'C20634',@createdate='2008-04-16 00:00:00:000',@status=N'Failed',@err=N'Attempted to read or write protected memory. This is often an indication that other memory is corrupt.

at SAPbobsCOM.IBusinessPartners.GetByKey(String CardCode)

at NetPoint.SynchSBO.SBOObjects.SBOPartner.SBOToNetPoint(SBOQueueObject qData)

at NetPoint.SynchSBO.SynchObjectBase.Synch()'

[Former Member · Accepted solution]

I'm probably going to jinx it for myself now, but I started a new initialize synch approx 3.5 hours ago and it is still running! (I think it is on the GL stage).

I changed the User ID setting to Seek and choose, and the password to webtools ID. These are the settings I changed to the last time I experienced the corrupt memory message.

Certainly watching the in SQL profiler shows a different pattern of commands/routines being run with no slow down, whereas before each time a customer record was being processed there would be a dramatic slow down.

It seems to me that different settings lead to different routines being run during the initialize synch (probably an obvious statement!), and that some of these routines are flawed in some way.

I have about another 44 hours to go before I will know if the synch is successful and whether or not the missing shipping details problem is solved.

[Former Member · Accepted solution]

Martin I tried contacting you this morning to do a webex, but it sounds like things are moving.

We tested this patch with your DB and had no crashes, just so you are aware.

[Former Member]

Hi Martin,

The only time I have seen anything like this was due to incorrect data in UDFs.

The case that caused this was a bit extreme, but I will relay it to you so you have some idea of how it might happen.

In the issue we experienced, an import tool was used that did not set data in required UDF fields (import was direct to the db). When the synch ran and these records were udpated, the synch crashed without any indication error.

I would try two things. First, look at the UDFs on the C21581D2 business partner. Make sure to pay attention to address and contact data as well. You might try updating the BP within B1 and see what happens. Second, If that doesn't resolve anything, put C21581D2 on skip in the synch and see if you can get past it.

One more long shot:

We had the previous error with B1 logging causing the OUT_OF_MEMORY error (which I thought you had resolved). You might double check your B1 config to make sure this is not rearing its head in another way.

The config is at C:\Program Files\SAP\SAP Business One\Log\b1LogConfig.xml by default.

The Trace attribute should be set to 0.

<b1logger Mode="A" MaxFileSize="5" MaxNumOfMsg="500" LogStack="0" Activate="1">
      <Components SystemMessage="1" SQLMessage="0" Trace="0" General="1" StockTool="0" Upgrade="0" Performance="0">
          <Component name="SystemMessage">
              <Severities Note="0" Warning="0" Error="1" CriticalError="1" BeProtective="1" AuditFailure="1">
                  <Severity name="BeProtective" LogStack="1" /> 
              </Severities>
          </Component>
      </Components>
  <Severities Note="0" Warning="0" Error="1" CriticalError="1" BeProtective="1" AuditFailure="1" /> 
  </b1logger>

[Former Member]

Hi Shane,

I have changed the trace attribute to 0 and am currently running another attempt. My previous attempt did result in out of memory errors being generated and synch manager crashing at a different point.

I think before this was to do with the contact user name and password settings. Certainly in my last run I changed the password generation to random to see if it would make any difference. I have now changed this back to webtools id.

I will let you know how I get on.