Information générale
Catégorie:
Base de données, Tables, Vues, Index et syntaxe SQL
Versions des environnements
Network:
Windows 2003 Server
Hi Frank,
>>Have you checked oplock setting ? Set SharingViolationDelay and SharingViolationTimeout to Zero on the server ?
>
>is the oplock setting you refer to the same as the SharingViolationRetries, SharingViolationDelay registry setting you refer to? Elsewhere I found that setting SharingViolationRetries to zero should help, so I am in the process of tryiing it out. I did not fimd anything on SharingViolationTimeout, did you mean Retries?
Yupp... memory is the first thing to go if too often exposed to google. You should find under SharingViolation and performance a few KB and other posts, sometimes about .mdb (access) file sharing.
OpLock is a caching behaviour useful for office documents but deemed hazardous for fileserver based data access. This "speeds up" the first user, but has to "unload" for the second user, thus adding to the speed difference. I gathered it to be safer to run fileserver data access always with oplocking turned off at lower speed, as there were reports of "malfunctions" running with oplocks on.
The settings for SharingViolation make things worse if I understood correctly, but you should watch SMB traffic if setting only delay to zero.
While we are at the topic: you might also check on the SMB protocol settings, there is a setting for SMB signing in Vista and some Server2003 versions nearly crippling performance.
I'ld appreciate, if you could post your findings with measurements <g>.
HTH
thomas
Précédent
Suivant
Répondre
Voir le fil de ce thread
Voir le fil de ce thread à partir de ce message seulement
Voir tous les messages de ce thread
Voir tous les messages de ce thread à partir de ce message seulement