Agree - given the differences between semantics with and without POSIX extensions we should document limitations that the CIFS client has in respect to locking including flock, posix locks and ofd locks. -- Best regards, Pavel Shilovsky ср, 24 февр. 2021 г. в 03:11, Aurélien Aptel <aaptel@xxxxxxxx>: > > Pavel Shilovsky <piastryyy@xxxxxxxxx> writes: > > If a flock is emulated on the server side with mandatory locks (which > > is what we only have for SMB2 without POSIX extensions) then we should > > maintain the same logic on the client. Otherwise you get different > > behavior depending on the caching policies currently in effect on the > > client side. You may consider testing with both modes when > > leases/oplocks are on and off. > > Hm.. you're right, the write will fail on the server side without > cache. > > I guess we should document current cifs behaviour in the flock man page. > > Cheers, > -- > Aurélien Aptel / SUSE Labs Samba Team > GPG: 1839 CB5F 9F5B FB9B AA97 8C99 03C8 A49B 521B D5D3 > SUSE Software Solutions Germany GmbH, Maxfeldstr. 5, 90409 Nürnberg, DE > GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 247165 (AG München) >
