[Orca-users] Re: Seeing Only one Disk
Phillip B. Bruce
pbbruce at mindspring.com
Sun Oct 21 11:17:51 PDT 2001
Blair Zajac wrote:
> "Phillip B. Bruce" wrote:
> >
> > Blair Zajac wrote:
> >
> > > "Phillip B. Bruce" wrote:
> > > >
> > > > Blair,
> > > >
> > > > The Disk Run Percent [Info] only shows c1t1d0 when I have
>
> > > lots
> > > > more than this.
> > > >
> > > > See Below:
> > > >
> > > > % df -k
> > > > Filesystem kbytes used avail capacity Mounted
> on
> > > > /dev/dsk/c0d0s0 288256 30817 228614 12% /
> > > > /dev/dsk/c0d0s6 1485664 527664 898574 37% /usr
> > > > /dev/dsk/c0d0p0:boot 10391 1623 8767 16% /boot
> > > > /proc 0 0 0 0% /proc
> > > > fd 0 0 0 0% /dev/fd
> > > > mnttab 0 0 0 0%
> /etc/mnttab
> > > > /dev/dsk/c0d0s4 240023 78361 137660 37% /var
> > > > swap 482280 4 482276 1% /var/run
> > > > swap 483368 1092 482276 1% /tmp
> > > > /dev/dsk/c0d0s3 650582 175588 416442 30% /opt
> > > > /dev/dsk/c0d1s0 2052750 1762916 228252 89% /u01
> > > > /dev/dsk/c0d1s3 2052750 979754 1011414 50% /web
> > > > /dev/dsk/c0d1s4 2052750 420300 1570868 22% /u02
> > > > /dev/dsk/c1d0s7 858015 778326 19628 98% /software
>
> > > > /dev/dsk/c0d1s1 2052750 865973 1125195 44% /opt/sfw
> > > > /dev/dsk/c0d0s7 960702 479060 424000 54%
> /export/home
> > > > /dev/dsk/c1d0s6 963183 910955 0 100%
> /usr/local
> > > >
> > > > While the others:
> > > > Disk System Wide Reads/Writes Per Second [Info]
> > > > Disk System Wide Transfer Rate [Info]
> > > > Disk Space Percent Usage [Info]
> > > > Disk Inode Percent Usage [Info]
> > > >
> > > > They report everything there but I'm skeptical of the Disk
> System
> > > Wide
> > > > Read/Writes Per Second
> > > > being correct or the System Wide Transfer. The Sparc Percent
> Usage
> > > and
> > > > Inode Percent Usage
> > > > is being reported correct. I should be seeing more. Is there a
> new
> > >
> > > > patch for this problem.
> > > >
> > > > I'm running this version
> > > > orca -v lib/orcallator.cfg
> > > > Orca version 0.264
> > > >
> > >
> > > Check your percol-* output files and see how many column names
> begin
> > > with
> > > disk_runp_. Is it just the one or are there more?
> > >
> > > You're running x86 Solaris, right? I don't see a c1t1d0 disk in
> your
> > > df -k output. Is this filesystem on another system?
> > >
> > > It looks like the regular expression for matching this is in
> > > orcallator.cfg
> > > does not match for these filesystems:
> > >
> > > data disk_runp_((?:c\d+t\d+d\d+)|(?:[ms]d\d+))
> > >
> > > It should be changed to something like
> > >
> > > data
> disk_runp_((?:c\d+t\d+d\d+)|(?:c\d+d\d+s\d+)|(?:[ms]d\d+))
> > >
> > > Try this and let me know if it works.
> > >
> > > The Disk System Wide Read/Writes Per Second and the System Wide
> > > Transfer
> > > rate could be wrong if not all of the disks are seen in the
> percol-*
> > > output files. However, if they are there, then these other
> > > measurements
> > > should be correct.
> > >
> > > I just logged into an x86 Solaris box and it had filesystems
> mounted
> > > like
> > > /dev/dsk/c0t0d0s3 1015874 244623 761093 25% /var
> > > which is the same for Sparc Solaris. Do you know why the device
> names
> > >
> > > are different from the x86 system I looked at?
> > >
> > > Blair
> > >
> >
> > Blair,
> >
> > I cut and paste the info you requested in the orcallator.cfg
> file.
> > I stop and restarted the orcallator and it still didn't get it.
> >
> > Keep in mind your x86 may be just all SCSI disks. That why it
> > captured the SCSI cdrom burner
> > on my system. The rest of the drives are IDE. That is why they
> look
> > funny in the df -k output.
> > Now you know why there isn't a SCSI target number.
> >
> > Here is my rrd output:
> >
> > ls *disk*
> > gauge_1024_X_disk_rK_per_s.rrd gauge_disk_runp_c1t1d0.rrd
> > gauge_1024_X_disk_wK_per_s.rrd gauge_disk_wr_per_s.rrd
> > gauge_disk_rd_per_s.rrd
> >
> > As you can see it still only see that c1t1d0 drive.
>
> Phillip,
>
> Check the percol-* files, not the RRD files, for all of the column
> names
> that contain disk_runp_*. The RRD files are only generated by Orca if
> Orca's
> configuration file finds the columns it is instructed to in the
> percol-*
> files.
>
> Blair
>
> --
> Blair Zajac <blair at orcaware.com> - Perl & sysadmin services for hire
> OS & web analysis - http://www.orcaware.com/orca/
>
> Yahoo! Groups Sponsor
[Image]
>
Blair,
It is finding those disk see below:
timestamp locltime DNnsrkcmdit uptime usr% sys% wio% idle% 1runq
5runq 15runq #proc #runque #waiting #swpque scanrate #proc/s
#proc/p5s smtx smtx/cpu ncpus mntC_/ mntU_/ mntA_/ mntP_/ mntc_/ mntu_/
mnta_/ mntp_/ mntC_/usr mntU_/usr mntA_/usr mntP_/usr mntc_/usr
mntu_/usr mnta_/usr mntp_/usr mntC_/var mntU_/var mntA_/var mntP_/var
mntc_/var mntu_/var mnta_/var mntp_/var mntC_/opt mntU_/opt mntA_/opt
mntP_/opt mntc_/opt mntu_/opt mnta_/opt mntp_/opt mntC_/u01 mntU_/u01
mntA_/u01 mntP_/u01 mntc_/u01 mntu_/u01 mnta_/u01 mntp_/u01 mntC_/web
mntU_/web mntA_/web mntP_/web mntc_/web mntu_/web mnta_/web mntp_/web
mntC_/u02 mntU_/u02 mntA_/u02 mntP_/u02 mntc_/u02 mntu_/u02 mnta_/u02
mntp_/u02 mntC_/software mntU_/software mntA_/software mntP_/software
mntc_/software mntu_/software mnta_/software mntp_/software
mntC_/opt/sfw mntU_/opt/sfw mntA_/opt/sfw mntP_/opt/sfw mntc_/opt/sfw
mntu_/opt/sfw mnta_/opt/sfw mntp_/opt/sfw mntC_/export/home
mntU_/export/home mntA_/export/home mntP_/export/home mntc_/export/home
mntu_/export/home mnta_/export/home mntp_/export/home mntC_/usr/local
mntU_/usr/local mntA_/usr/local mntP_/usr/local mntc_/usr/local
mntu_/usr/local mnta_/usr/local mntp_/usr/local disk_runp_c0d0
disk_runp_c0d1 disk_runp_c1d0 disk_runp_c1t1d0 disk_peak disk_mean
disk_rd/s disk_wr/s disk_rK/s disk_wK/s swap_avail page_rstim freememK
free_pages dnet0Ipkt/s dnet0Opkt/s dnet0InKB/s dnet0OuKB/s dnet0IErr/s
dnet0OErr/s dnet0Coll% dnet0NoCP/s dnet0Defr/s tcp_Iseg/s tcp_Oseg/s
tcp_InKB/s tcp_OuKB/s tcp_Ret% tcp_Dup% tcp_Icn/s tcp_Ocn/s tcp_estb
tcp_Rst/s tcp_Atf/s tcp_Ldrp/s tcp_LdQ0/s tcp_HOdp/s nfs_call/s
nfs_timo/s nfs_badx/s nfss_calls nfss_bad v2reads v2writes v3reads
v3writes dnlc_ref/s dnlc_hit% inod_ref/s inod_hit% inod_stl/s pp_kernel
pagesfree pageslock pagesio pagestotl #httpds httpop/s http/p5s cndget/s
search/s cgi/s htErr/s httpb/s %to1KB %to10KB %to100KB %to1MB
%over1MB NoGatway
Then why isn't it graphing the data then. If it is finding it?
--
************************************************************
*** Phillip B. Bruce ***
*** http://pbbruce.home.mindspring.com ***
*** pbbruce at mindspring.com ***
*** ***
*** "Have you ever noticed? Anybody going slower than ***
*** you is an idiot, and anyone going faster than you ***
*** is a maniac." - George Carlin ***
************************************************************
More information about the Orca-users
mailing list