Date: Fri, 3 Jul 1998 17:45:50 +0900 (KST) From: swjeong <swjeong@net.kitel.co.kr> To: freebsd-scsi@FreeBSD.ORG Subject: Re: [Q] double-ending SCSI RESET problem Message-ID: <199807030845.RAA15364@net.kitel.co.kr>
next in thread | raw e-mail | index | archive | help
# Environment
- I am using CAM ('3.0-980520-SNAP + CAM-980520') as follows.
AHA3940UW AHA3940UW
SCSC-adapter HDD HDD HDD .... HDD SCSI-adapter
| | | | | |
+------+----------------+------+------+---------+----------+--------+
| |
Term. Term.
(3940's Termination is disabled and scsi bus is connected on internal
channel B, channel A is used for boot disk)
# Problem discussed up to now.
After machine A boot, I booted machine B.
In the middle of kernel-autoconfiguration-stage,
many many 'Someone reset channel A' messages occurs on machine A and B.
Tt lasts from few seconds to forever.
Then B panics, if with luck, B remains alive.
And made kernel configuration file correct
as "Justin T. Gibbs" said.
> controller ahc0
> controller ahc1
> controller scbus0 at ahc0
> controller scbus1 at ahc1
But, the problem persisted.
# Question.
I applied following two awkward patch to escape above problem.
aic7xxx.c patch makes ahc not to issue RESET again,
when it detected RESET on the bus.
scsi_all.c patch makes opearation not to stopped by
UNIT ATTENTION.
After this patch, 'Someone reset channel A' doen't
occurs continuously. and disk operation is not halted by
peer's rebooting.
Does this patch make sense ?
$ diff aic7xxx.c.orig aic7xxx.c
4140c4140
< ahc_reset_current_bus(ahc);
---
> /* ahc_reset_current_bus(ahc); */
4152c4152
< ahc_reset_current_bus(ahc);
---
> /* ahc_reset_current_bus(ahc); */
$ diff scsi_all.c.orig scsi_all.c
2134a2135,2139
>
> error = ERESTART;
> print_sense = FALSE;
> break;
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199807030845.RAA15364>