[Orca-dev] Raw disks (Fiber channel disks) causing Segmentation Faults in 'orcallator.se'

[B.Schopp at gmx.de]

Hi,

> Got the latest version of Orca and SE toolkit (3.3)
> running on Solaris 8 (SF3800's with Sun's T3 storage
> and Sun Cluster 3.x/Veritas Volume Mgr.).   Am getting
> segmentation faults when starting SE Toolkit with
> default 'orcallator.se' RAW_DISK values.  Have been
> able to trace it down to a problem with strncmp()
> which occurs when handling kstat structures of raw
> disks. (For now, I've just uncommented out the
> #define USE_RAWDISK     1
> parameter).

I have got the same problem over here: 4 servers connected
to T3s in clusterconfiguration are not capable tto run orcallator.
I get also the same effect with commenting out USE_RAWDISK.

> I am curious as to whether someone has already worked
> on a modified version of an orcallator.se to be able
> to capture RAW disk statistics on fiber channel disks.

Seems like a good chance to me to get back and into the deep
of C programming... ;-) But it will take a while, as actually other
projects on work are more important.  Maybe i get the permission
to work on it with a colleage or two, if i can escalate the error as
to important for our data-gathering.

Greetings,

Burkhardt

--
Daily bofh-style excuse:
NOTICE: alloc: /dev/null: filesystem full

-- 
+++ GMX - Mail, Messaging & more  http://www.gmx.net +++
Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage!