--- On Wed, 18 Mar 1998 22:29:40 -0500 Alexei Rodriguez alexei@cimedia.com wrote:
This sounds very odd. We used NetApp's for news at my previous job and they worked beautifully.
What OS and patch levels do you have on the Sun. Not that SS20's are real speed demons...
Sparc 20 upgraded to dual ross 150's. 512meg ram 2 fast 2.1 gig internals 2 fastwide 4 gig externals.
Release: 5.6 Kernel architecture: sun4m Application architecture: sparc Hardware provider: Sun_Microsystems Domain: Kernel version: SunOS 5.6 Generic 105181-04 January 1998
What version of DataONTAP do you have on the filer? What ndd setting
have
you applied to the SS20?
NetApp Release 5.0, upgraded a couple of days ago and problem didnt change. We were running 4.3.1 before this. Are there any nfs
/usr/sbin/ndd -set /dev/ip ip_respond_to_echo_broadcast 0 /usr/sbin/ndd -set /dev/ip ip_enable_group_ifs 0 /usr/sbin/ndd -set /dev/tcp tcp_recv_hiwat 65536 /usr/sbin/ndd -set /dev/tcp tcp_xmit_hiwat 65536 /usr/sbin/ndd -set /dev/tcp tcp_cwnd_max 65535 /usr/sbin/ndd -set /dev/tcp tcp_ip_abort_cinterval 30000 /usr/sbin/ndd -set /dev/tcp tcp_close_wait_interval 30000
Doe the large writes correspond to a specific news hierarchy (ie:
misc.jobs.*
or alt.binaries.*)?
Any file over 30k or so.
What does sysstat look like when this occurs?
CPU NFS CIFS HTTP Net kB/s Disk kB/s Tape kB/s Cache in out read write read write age 0% 48 0 0 26 11 88 0 0 0 5 0% 101 0 0 50 20 296 0 0 0 5 1% 104 0 0 54 32 148 0 0 0 5 2% 117 0 0 70 24 368 0 0 0 5 1% 80 0 0 41 12 140 0 0 0 5 3% 56 0 0 33 11 248 0 0 0 5 2% 71 0 0 44 9 356 0 0 0 5 7% 111 0 0 67 18 2428 1332 0 0 5 19% 48 0 0 18 6 4808 5284 0 0 5 4% 86 0 0 43 11 680 1376 0 0 5 0% 36 0 0 12 143 272 0 0 0 5 1% 104 0 0 46 29 484 0 0 0 5 2% 69 0 0 36 14 120 0 0 0 5 2% 73 0 0 39 14 52 0 0 0 5 1% 84 0 0 41 13 180 0 0 0 5 2% 65 0 0 39 8 228 0 0 0 5 2% 61 0 0 37 11 152 0 0 0 5 7% 143 0 0 795 23 44 0 0 0 5
Alexei
Thanks -----------------End of Original Message-----------------
David Power V-P Operations Insync Internet Services Inc.