Workstation engine 6.15 and Windows NT Server

[ajuelich]

Our application accesses over the workstation engine the data. The data are stored on the NT server. File sharing is used. Under certain circumstances common accesses to a file cause the errors 2 and 4. If the data are stored on a Win 98 Client, then the error is not comprehensible. I assume, which " shadow pages " under NT are not administered correctly. Who can help? Ajuelich

 

[Pervasivite]

Status 2 and 4 are most commonly caused by bad hardware or problems with the OS software. It would take some testing to determine what the root cause of the Status codes were, but there can be file locking issues with high volume database files accesses inherent in NT when you get above 5 or 10 users, so that is where it starts to pay off in using the client/server version. If it were me having these issues I might try downloading the Pervasive.SQL 2000i server version which comes with a 20 user 30 day evaluation key and see if that took care of the issues. If not you have hardare corruption issues at the server side. It would be helpful to know how many users you have running concurrently and what application you are using with 6.15? It should plug into Pervasive.SQL 2000i which has a complete Btrieve 7.9 engine in it that is completely backwards compatible and natively supports all Btrieve applications and data. Also exactly which version of Btrieve 6.15 are you using? I doubt the shadow paging is getting messed up unless your systems are subject to frequent unexpected power outages. There was one known bug in that case and there is an update to the 32-bit Windows Btrieve 6.15 engine if that is what you are using.

Pervasivite

 

[nihsumi]

Hi Ajuelich,
In addition to what Pervasivite has written:
Status 2s and 4s are so different I was wondering are you getting them on the same Btrieve call and to the same database?

I agree w/Perv about status 2 problems and would add that I would rebuild any file getting status 2s before using it again, perhaps on a different file system/drive.

If after you have rebuilt the file and verified good hardware and are still seeing the status 2s w/4s on the same file and same Btrieve call I would start thinking about my RAM. Also you might try increasing your Pervasive cache. Varying Pevasive's flush times and operations my lead you to the culprit.

 

[ajuelich]

Hi friends,
after changing the parameter for opportunistic locking on the server and the clients we have no problems.
REGEDIT4

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\LanmanServer\Parameters]
"EnableOpLocks"=dword:00000000
"EnableOpLockForceClose"=dword:00000001
"CachedOpenLimit"=dword:00000000

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\LanmanWorkstation\Parameters]
"UseOpportunisticLocking"=dword:00000000
"UtilizeNtCaching"=dword:00000000



Thanks

ajuelich