I started with a test file some 60+M big (/tmp/TEST), and did the following:
dd if=/tmp/TEST of=/nfs/u22/tmp/1 bs=(filesize of TEST) &
dd if=/tmp/TEST of=/nfs/u22/tmp/2 bs=(filesize of TEST) &
dd if=/tmp/TEST of=/nfs/u22/tmp/3 bs=(filesize of TEST) &
dd if=/tmp/TEST of=/nfs/u22/tmp/4 bs=(filesize of TEST) &
dd if=/tmp/TEST of=/nfs/u22/tmp/5 bs=(filesize of TEST) &
It writes 188416 bytes to /nfs/u22/tmp/1, 184320 bytes to
/nfs/u22/tmp/2, none to the other two. Do a single file at a time
doesn't lock it. This is a VERY really concern as we regularly dump
file in the 80M range, often several at a time. (Speech Recog. is a
bitch on the systems).
The machine is 128M, Pentium 75+, stepping 12, plain vanilla 2.1.119
UP. Any other details can be furnished upon request. The machine being
written to is a Solaris 2.5.1 server which is pretty close to patch
level. knfsd on the linux box is 0.4.22 (I doubt that matters here
though). I'm open to running patches, etc. if you think it'll help.
The lock is absolutely silent, nothing logs, nothing on the
console. Distro is RH 5.1, Compiled with gcc 2.7.2.3. (Anything I
missed?)
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Justin Hahn <jehahn@raven.bu.edu>
Systems Administrator, Boston University SPI Lab
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html