[Fwd: lost SCSI IO]

[Fwd: lost SCSI IO]

[NucleoDyne Systems, Inc.]

[-- Attachment #1: Type: text/plain, Size: 73 bytes --]


-- 
--
NucleoDyne Corporation
nucleon@nucleodyne.com
www.nucleodyne.com

[-- Attachment #2: Forwarded message - lost SCSI IO --]
[-- Type: message/rfc822, Size: 3229 bytes --]

From: "NucleoDyne Systems Inc." <nucleon@nucleodyne.com>
To: linux-kernel@vger.kernel.org
Subject: lost SCSI IO
Date: Mon, 08 Dec 2003 11:29:07 -0800
Message-ID: <3FD4D103.2080007@nucleodyne.com>

I have been developing a SCSI based application and facing hardship due 
to lack of  better debugging support in SCSI domain on linux. May be I 
am not very familiar with linux scsi subsystem.

 I have a lost IO, sitting somewhere in some queue. The scsi logging 
facility has been turned on with : echo "scsi log all" > /proc/scsi/scsi.

The syslog shows that the request has started:
scsi_do_req (host = 0, channel = 0 target = 3, buffer =00000000, bufflen 
= 0, d)command : 00  20  00  00  00  00
Leaving scsi_do_req()

The SCSI bus trace does not show any activity.

I guess only way to find out the state of the IO is to put printf and 
recompile the kernel. HP-UX had facilities like  lkcd, linux crash dump 
analyzer. I used to be called q4. A perl script could be written to 
navigate kernel data structures and extract information from them on a 
running system. If we had that kind of tool already into linux then 
debugging the live system would be easier.

Is there a plan to include lkcd into  default kernel?



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