Hi
netiosrv.prg - as is
RUN: "./netiosrv_linux 60001 192.168.1.102"
netiotst.prg:
define DBNAME "net:data/_tst_"
request DBFCDX
proc main()
set exclusive off
rddSetDefault( "DBFCDX" )
? "NETIO_CONNECT():", netio_connect( "192.168.1.102", 60001 )
if ! dbExists( DBNAME )
createdb( DBNAME )
endif
testdb( DBNAME )
return
proc createdb( cName )
dbCreate( cName, {{"F1", "C", 20, 0}} )
use (cName)
dbAppend()
index on field->F1 tag T1 //without it - all right
close
return
proc testdb( cName )
use (cName)
dbgoto(1)
? "DbrLock: second run (in parallel) - must be an error block"
dbrlock()
field->f1 := "test"
? "Locked"
wait
close
return
2009/9/17 Przemyslaw Czerpak <[email protected]>:
> On Thu, 17 Sep 2009, Alexandr Okhotnikov wrote:
>
> Hi,
>
>> I tested it on linux:
>> lock through NETIO not work (rewriting a locked record, and then flies
>> out to destroy the index).
>
> Locks works perfectly.
>
>> removing the "net:" all earned
>> What could be wrong?
>
> If you can create as small as possible peace of code to replicate
> the situation which you think is a problem for you then maybe I can
> give you an answer. Now I do not know what you are doing and what
> you expect. Your sentence above suggests that you expect that Linux
> kernels will work with mandatory locking and DOS like deny flags but
> it's highly possible that I didn't guess correctly.
>
> best regards,
> Przemek
_______________________________________________
Harbour mailing list
[email protected]
http://lists.harbour-project.org/mailman/listinfo/harbour
