[4244] in linux-scsi channel archive
Re: SCSI Boot Problem with 2.0.34 SMP
daemon@ATHENA.MIT.EDU (Brendan Miller)
Sun Jun 28 02:58:43 1998
Date: Sat, 27 Jun 1998 23:49:27 -0700
From: Brendan Miller <brmiller@xenon.kudzu.gammalink.com>
To: "Frederick F. Gleason" <fredg@wava.com>, linux-scsi@vger.rutgers.edu,
linux-smp@rutgers.edu, necaise@home.com
Cc: Brendan Miller <brmiller@wco.com>
Reply-To: Brendan Miller <brmiller@wco.com>
In-Reply-To: <Pine.LNX.3.95.980626040951.28169A-100000@clarke.envise.com>; from Christopher B. Smith on Fri, Jun 26, 1998 at 04:12:54AM -0400
Well, after about 20 reboots with different kernels today, I come up with
the following:
2.0.33 UP - aic7xxx, 2940, and 2 HP drives okay
2.0.33 SMP - aic7xxx, 2940, and 2 HP drives okay
2.0.34 UP - aic7xxx (not patched), 2940, and 2 HP drives: "Bad scbptr during
SELTO", SCSISEQ, SEQADDR, SSTAT0 error
2.0.34 SMP - aic7xxx (not patched), 2940, and 2 HP drives: "Bad scbptr during
SELTO", SCSISEQ, SEQADDR, SSTAT0 error
2.0.34 UP - aic7xxx (patched with "aic7xxx-5.0.19-2.0.34.patch"), 2940, and
2 HP drives: "Bad scbptr during SELTO", SCSISEQ, SEQADDR,
SSTAT0 error
2.0.34 SMP - aic7xxx (patched with "aic7xxx-5.0.19-2.0.34.patch"), 2940, and
2 HP drives: "Bad scbptr during SELTO", SCSISEQ, SEQADDR,
SSTAT0 error
2.0.34 SMP - aic7xxx (patched with "aic7xxx-5.0.19-2.0.34.patch"), 2940UW, and
2 HP drives: completely failed to detect drives on bus
2.1.105 SMP - aic7xxx (unpatched), 2940, and 2 HP drives: rarely detects
both drives--sometimes the "Bad scbptr" error, othertimes,
just fails to see both drives altogether
2.1.105 SMP - aic7xxx (unpatched), 2940, and 2 HP drives: completely failed
to detect drives on bus
So obviously, 2.0.34 is out. Can someone comment on the aic7xxx driver,
either as part of 2.0.34, or the aic7xxx-5.0.19-2.0.34.patch (which didn't
seem to help me)? Moreover, I *really* like the IOAPIC stuff of 2.1.x.
But, I couldn't get 2.1.105 to reliably detect both drives on the bus.
I wasn't gonna try 2.1.106 (didn't someone say it was really hosed?), and
my /usr/src partition was out of room, so I had to reorganize my disk
space, and haven't built 2.1.107 yet.
Another thing. You may recall from my previous posts that I had a ZIP
drive on this chain. (Also, earlier, I only had one HP drive--I had
another in the static-bag waiting to "get on the bus".) Anyway, I took
the ZIP drive off the chain today, as I would get the following (when
booting 2.0.33 SMP and having the aic7xxx do it's thing:
scsi1: Scanning channel A for devices.
Vendor: HP Model: C3725S Rev: 6039
Type: Direct-Access ANSI SCSI revision: 02
Detected scsi disk sdd at scsi1, channel 0, id 1, lun 0
Vendor: HP Model: C3725S Rev: 6039
Type: Direct-Access ANSI SCSI revision: 02
Detected scsi disk sde at scsi1, channel 0, id 2, lun 0
Vendor: IOMEGA Model: ZIP 100 Rev: D.09
Type: Direct-Access ANSI SCSI revision: 02
Detected scsi removable disk sdf at scsi1, channel 0, id 5, lun 0
SCSI device sdd: hdwr sector= 512 bytes. Sectors= 4194058 [2047 MB] [2.0 GB]
sdd: sdd1 sdd2 sdd3
SCSI device sde: hdwr sector= 512 bytes. Sectors= 4194058 [2047 MB] [2.0 GB]
sde: sde1 sde2
You notice that in the bus scan, it locates all thre drives, but doesn't
list the partition table for the ZIP drive. There is no /dev/sdf!! I
mean, the special file is there, but "fdisk /dev/sdf" does not work
(uanble to open /dev/sdf message). It's really weird. The two HP drives
are in an external SCSI enclosure, the ZIP drive is attached to it with
a Centronics to 25pin SCSI, and the ZIP drive provides termination. I'm
beginning to wonder if I should put my trusty AIC-6360 card back in to
keep that 25pin SCSI funk off the 2940 bus. I originally took it off
my BusLogic BT-958 bus, because I feared it would cause problems.
Anyway, the point of this was to let everyone know where I stood with
my 2940 problems. I have all my "primary" disks (/, /usr, /usr/local,
/home, etc.) on disks connected to a BT-958. The two HPs on the 2940
are just for archival purposes. I was gonna play with mirroring some
day.
Brendan Miller
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
