[Orca-users] Fwd: orcallator segmentation fault

[linuxrules at hushmail.com]

I'm having an issue with orcallator running on a 12k domain that is connected
to a Hitachi storage system. I beleive it is probably the issue related
to RAW_disk issues that others have brought up on this list before. I
tried increasing the MAX_RAWDISKS and MAX_COLUMNS in orcallator.se with
no luck. Running '/opt/RICHPse/bin/se -DWATCH_OS -d orcallator.se' generates
a lot of output ending with this.

total_disk_reads += RAW_disk[31].reads<0>
total_disk_writes += RAW_disk[31].writes<0>
total_disk_readk += RAW_disk[31].kreads<0>
total_disk_writek += RAW_disk[31].kwrites<0>
mean_disk_busy += RAW_disk[31].run_percent<0>
if (RAW_disk[31].run_percent<0> > peak_disk_busy<0.427329>)
i++;
if (RAW_disk[32].short_name[0]<s> == <115> && RAW_disk[32].short_name[1]<d>
== <116>)
if (RAW_disk[32].short_name[0]<s> == <102> && RAW_disk[32].short_name[1]<d>
== <100>)
disk_count++;
put_output(sprintf(<disk_runp_%s>, RAW_disk[32].long_name[32]<s>), sprintf(<%16.5f>,

 RAW_disk[32].run_percent<0>))
if (current_column<232> >= <4096>)
col_comment[232] = comment<disk_runp_ssd29.fp4>
col_data[232] = data<         0.00000>
current_column++;
total_disk_reads += RAW_disk[32].reads<0>
total_disk_writes += RAW_disk[32].writes<0>
total_disk_readk += RAW_disk[32].kreads<0>
total_disk_writek += RAW_disk[32].kwrites<0>
mean_disk_busy += RAW_disk[32].run_percent<0>
if (RAW_disk[32].run_percent<0> > peak_disk_busy<0.427329>)
i++;
if (RAW_disk[33].short_name[0]<s> == <115> && RAW_disk[33].short_name[1]<d>
== <116>)
if (RAW_disk[33].short_name[0]<s> == <102> && RAW_disk[33].short_name[1]<d>
== <100>)
disk_count++;
put_output(sprintf(<disk_runp_%s>, RAW_disk[33].long_name[33]<d>), sprintf(<%16.5f>,

 RAW_disk[33].run_percent<0>))
if (current_column<233> >= <4096>)
col_comment[233] = comment<disk_runp_ssd30.fp4>
col_data[233] = data<         0.00000>
current_column++;
total_disk_reads += RAW_disk[33].reads<0>
Segmentation Fault

I have read in a previous post that "odd" filenames in /dev/dsk and /dev/rdsk
could cause this issue. I do have quite a few files in their that aren't
exactly normal. I believe these are the disks that are in the Hitachi
storage system. Here is a look at some of the filenames in /dev/dsk:

lrwxrwxrwx   1 root     root          63 Oct 23 21:31 c10t500060E80000000000007D
F00000040Fd0s0 -> ../../devices/scsi_vhci/ssd at g500060e80000000000007df00000040f:
a
lrwxrwxrwx   1 root     root          63 Oct 23 21:31 c10t500060E80000000000007D
F00000040Fd0s1 -> ../../devices/scsi_vhci/ssd at g500060e80000000000007df00000040f:
b
lrwxrwxrwx   1 root     root          63 Oct 23 21:31 c10t500060E80000000000007D
F00000040Fd0s2 -> ../../devices/scsi_vhci/ssd at g500060e80000000000007df00000040f:
c

Here is a sample from /dev/rdsk:

lrwxrwxrwx   1 root     root          67 Oct 23 21:31 c10t500060E80000000000007D
F00000040Fd0s0 -> ../../devices/scsi_vhci/ssd at g500060e80000000000007df00000040f:
a,raw
lrwxrwxrwx   1 root     root          67 Oct 23 21:31 c10t500060E80000000000007D
F00000040Fd0s1 -> ../../devices/scsi_vhci/ssd at g500060e80000000000007df00000040f:
b,raw
lrwxrwxrwx   1 root     root          67 Oct 23 21:31 c10t500060E80000000000007D
F00000040Fd0s2 -> ../../devices/scsi_vhci/ssd at g500060e80000000000007df00000040f:
c,raw


Is there a workaround  that will allow me to gather performance data
for my disks? I have heard that if you comment out the USE_RAWDISK line
from the orcallator.se file that this will stop orcallator from crashing
but you lose the ability to gather performance data for your disks.

Thanks in advance for the help.

Ed

PS - I'm using se version 3.2.1 and orcallator.se version 1.37










Concerned about your privacy? Follow this link to get
FREE encrypted email: https://www.hushmail.com/?l=2

Free, ultra-private instant messaging with Hush Messenger
https://www.hushmail.com/services.php?subloc=messenger&l=434

Promote security and make money with the Hushmail Affiliate Program: 
https://www.hushmail.com/about.php?subloc=affiliate&l=427