people are familiar with the temporary performance hole in one's toaster when starting a backup via "rsh toaster dump 0fqwerty etc. etc.", right?
am i imagining it, or is it getting worse with later versions of ONTAP? it's got to the point where users are substantially hampered in working over the weekend when the backups kick off becasue CPU goes to 100% used and stays there for tens of minutes (and, for that matter, at various times during the backups). am i doing something wrong?
Tom Yates - Unix Chap - The Mathworks, Inc. - +1 (508) 647 7561 MAG#65061 DoD#0135 AMA#461546 1024/CFDFDE39 0C E7 46 60 BB 96 87 05 04 BD FB F8 BB 20 C1 8C
We have what seems to be a reproducible problem in trying to do something like a du or find in a certain directory on our filer, from a client running Solaris 2.5.1. That client mount home directories under /home, and attempting to do a du or find on /home/rts generates
...lot 'o stuff open("convert", O_RDONLY|O_NDELAY) = 12 fcntl(12, F_SETFD, 0x00000001) = 0 fstat(12, 0xEFFFEA40) = 0 write(1, " . / p / P r o j e c t s".., 81) = 81 chdir("convert/") getdents(12, 0x00027A50, 1048) (sleeping...) NFS server sinagua not responding still trying
The full pathname of that directory appears to be
/home/rts/p/Projects/Scorpion/distributions/6.1/scorpion6.1/examples/functionlang/convert
The filer is an F520 running 4.3.1D7. Nothing usual is in the messages log, and "sysstat 1" isn't showing anything unusual on the filer. The Solaris system is running with most (maybe not quite all) of the latest recommended patches.
-
Jim Davis -
Tom Yates