aacraid, dbus, and RAID health status

aacraid, dbus, and RAID health status

[Torrey Hoffman]

Hi.

I'm interested in setting up a system which can monitor the status of an
aacraid-based RAID system, ideally via dbus.

I've seen solutions on the web that poll the system using the 'afacli'
command line tool, and parse the output.  This seems less than optimal.

I want a system which doesn't use polling.  Surely it's possible to have
the kernel send out notifications if the health status of the RAID
changes?

Ideally these would be available via HAL (hald) and sent to user space
via dbus.  

I'm willing to patch the kernel and hald if required.  If no such
solution exists, I may be able to spend some paid work time writing it.
If I did, would there be some chance of getting it into the official
driver?  Would other people find this useful? 

I've been grepping the sources of .../drivers/scsi/... looking for
examples of kernel events or hotplug being used to notify user space of
changes in the RAID array health, without any luck.  Does any other
driver, (or the software RAID system) do anything like this? 

Thanks in advance for any tips or pointers. 

Please reply directly, as I'm not currently subscribed to linux-scsi. If
I get multiple off-list messages, I'll write and post a summary. 

Torrey Hoffman
torrey@lockdownnetworks.com
thoffman@arnor.net

RE: aacraid, dbus, and RAID health status

[Salyzyn, Mark]

[-- Attachment #1: Type: text/plain, Size: 2495 bytes --]

You will find this GPL application will help, and depending on the card,
will have varying levels of details. Just change the code that reports
to the syslog daemon for your own purposes. This tool looks at the AIF
delivery and reports internationalized UTF8 strings to the syslog daemon
on critical as well as minor changes in component status.

You will also see that the Adaptec version of the driver (available on
our tech support web pages, or directly from me upon request for
latest/greatest) does sniffing of these same AIFs and provides hot-plug
events based on critical array status changes. This 'hot-plug' code is
in the queue to eventually land on the scsi list...

Sincerely -- Mark Salyzyn


-----Original Message-----
From: linux-scsi-owner@vger.kernel.org
[mailto:linux-scsi-owner@vger.kernel.org] On Behalf Of Torrey Hoffman
Sent: Tuesday, August 16, 2005 12:18 PM
To: linux-scsi@vger.kernel.org
Subject: aacraid, dbus, and RAID health status


Hi.

I'm interested in setting up a system which can monitor the status of an
aacraid-based RAID system, ideally via dbus.

I've seen solutions on the web that poll the system using the 'afacli'
command line tool, and parse the output.  This seems less than optimal.

I want a system which doesn't use polling.  Surely it's possible to have
the kernel send out notifications if the health status of the RAID
changes?

Ideally these would be available via HAL (hald) and sent to user space
via dbus.  

I'm willing to patch the kernel and hald if required.  If no such
solution exists, I may be able to spend some paid work time writing it.
If I did, would there be some chance of getting it into the official
driver?  Would other people find this useful? 

I've been grepping the sources of .../drivers/scsi/... looking for
examples of kernel events or hotplug being used to notify user space of
changes in the RAID array health, without any luck.  Does any other
driver, (or the software RAID system) do anything like this? 

Thanks in advance for any tips or pointers. 

Please reply directly, as I'm not currently subscribed to linux-scsi. If
I get multiple off-list messages, I'll write and post a summary. 

Torrey Hoffman
torrey@lockdownnetworks.com
thoffman@arnor.net


-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.kernel.org More majordomo info
at  http://vger.kernel.org/majordomo-info.html

[-- Attachment #2: aeventd-1.0-4.i386.rpm --]
[-- Type: application/octet-stream, Size: 64825 bytes --]