I am running a pretty standard RH7.3 configuration (Kernel 2.4.18)
with root and key filesystems on a 20Gig EIDE disk.
I also have some older ext2 partitions (circa Kernel 2.0) on a 2Gig
SCSI running on an Adaptac 2490 adapter.
When I mount the SCSI ext2 partitions, everything works fine for a
while but then access to the partition freezes for several minutes
associated with the thousands of scary syslog messages (see below).
SCSI termination is done properly.
Also, it should be noted that the SCSI drive is *rock* solid when
booted under an old 2.0.28 kernel or when accessed from Win95. In
fact, just before the most recent incident, I had fuly fscked the old
Linux partitions (including the 'badblocks' option) under 2.0.28.
Any suggestions?
** Could you please copy me on an email response, since I do not (yet)
subscribe to the mailing list? . Thanks very much for your help.
--------
syslog extacts follow:
Sep 2 11:42:09 surgeon kernel: attempt to access beyond end of device
Sep 2 11:42:09 surgeon kernel: 08:02: rw=0, want=7053826, limit=32130
Sep 2 11:42:09 surgeon kernel: attempt to access beyond end of device
Sep 2 11:42:09 surgeon kernel: 08:02: rw=0, want=7054081, limit=32130
....
... (plus *thousands* more lines like this)
....
Sep 2 12:01:17 surgeon kernel: scsi0:0:0:0: Attempting to queue an ABORT message
Sep 2 12:01:17 surgeon kernel: scsi0: Dumping Card State in Data-in phase, at SEQADDR 0x79
Sep 2 12:01:17 surgeon kernel: ACCUM = 0x0, SINDEX = 0x8, DINDEX = 0x8f, ARG_2 = 0xff
Sep 2 12:01:17 surgeon kernel: HCNT = 0x18 SCBPTR = 0xd
Sep 2 12:01:17 surgeon kernel: SCSISEQ = 0x12, SBLKCTL = 0x0
Sep 2 12:01:17 surgeon kernel: DFCNTRL = 0x78, DFSTATUS = 0x0
Sep 2 12:01:17 surgeon kernel: LASTPHASE = 0x40, SCSISIGI = 0x44, SXFRCTL0 = 0x80
Sep 2 12:01:17 surgeon kernel: SSTAT0 = 0x0, SSTAT1 = 0x2
Sep 2 12:01:17 surgeon kernel: STACK == 0x96, 0x186, 0x156, 0x0
Sep 2 12:01:17 surgeon kernel: SCB count = 60
Sep 2 12:01:17 surgeon kernel: Kernel NEXTQSCB = 34
Sep 2 12:01:17 surgeon kernel: Card NEXTQSCB = 39
Sep 2 12:01:17 surgeon kernel: QINFIFO entries: 39 31 45 9 52 46 54 50 10 33 28 29 35 32 43 47 40 19 13 6 55
Sep 2 12:01:17 surgeon kernel: Waiting Queue entries:
Sep 2 12:01:17 surgeon kernel: Disconnected Queue entries:
Sep 2 12:01:17 surgeon kernel: QOUTFIFO entries:
Sep 2 12:01:17 surgeon kernel: Sequencer Free SCB List: 3 7 1 11 4 0 8 9 15 12 2 14 6 5 10
Sep 2 12:01:17 surgeon kernel: Sequencer SCB Info: 0(c 0x60, s 0x7, l 0, t 0xff) 1(c 0x60, s 0x7, l 0, t 0xff) 2(c 0x60, s 0x7, l 0, t 0xff) 3(c 0x60, s 0x7, l 0, t 0xff) 4(c 0x60, s 0x7, l 0, t 0xff) 5(c 0x60, s 0x7, l 0, t 0xff) 6(c 0x60, s 0x7, l 0, t 0xff) 7(c 0x60, s 0x7, l 0, t 0xff) 8(c 0x60, s 0x7, l 0, t 0xff) 9(c 0x60, s 0x7, l 0, t 0xff) 10(c 0x60, s 0x7, l 0, t 0xff) 11(c 0x60, s 0x7, l 0, t 0xff) 12(c 0x60, s 0x7, l 0, t 0xff) 13(c 0x60, s 0x7, l 0, t 0x29) 14(c 0x60, s 0x7, l 0, t 0xff) 15(c 0x60, s 0x7, l 0, t 0xff)
Sep 2 12:01:17 surgeon kernel: Pending list: 55(c 0x60, s 0x7, l 0), 6(c 0x60, s 0x7, l 0), 13(c 0x60, s 0x7, l 0), 19(c 0x60, s 0x7, l 0), 40(c 0x60, s 0x7, l 0), 47(c 0x60, s 0x7, l 0), 43(c 0x60, s 0x7, l 0), 32(c 0x60, s 0x7, l 0), 35(c 0x60, s 0x7, l 0), 29(c 0x60, s 0x7, l 0), 28(c 0x60, s 0x7, l 0), 33(c 0x60, s 0x7, l 0), 10(c 0x60, s 0x7, l 0), 50(c 0x60, s 0x7, l 0), 54(c 0x60, s 0x7, l 0), 46(c 0x60, s 0x7, l 0), 52(c 0x60, s 0x7, l 0), 9(c 0x60, s 0x7, l 0), 45(c 0x60, s 0x7, l 0), 31(c 0x60, s 0x7, l 0), 39(c 0x60, s 0x7, l 0), 41(c 0x60, s 0x7, l 0)
Sep 2 12:01:17 surgeon kernel: Kernel Free SCB list: 1 44 48 51 36 37 38 49 42 11 27 2 0 16 3 12 25 17 53 59 58 22 23 4 8 24 57 7 30 26 5 21 15 14 20 18 56
Sep 2 12:01:17 surgeon kernel: DevQ(0:0:0): 0 waiting
Sep 2 12:01:17 surgeon kernel: DevQ(0:5:0): 0 waiting
Sep 2 12:01:17 surgeon kernel: DevQ(0:6:0): 0 waiting
Sep 2 12:01:17 surgeon kernel: scsi0:0:0:0: Cmd aborted from QINFIFO
Sep 2 12:01:17 surgeon kernel: aic7xxx_abort returns 0x2002
Sep 2 12:01:27 surgeon kernel: scsi0:0:0:0: Attempting to queue an ABORT message
Sep 2 12:01:27 surgeon kernel: scsi0: Dumping Card State in Data-in phase, at SEQADDR 0x7a
Sep 2 12:01:27 surgeon kernel: ACCUM = 0x0, SINDEX = 0x8, DINDEX = 0x8f, ARG_2 = 0xff
Sep 2 12:01:27 surgeon kernel: HCNT = 0x18 SCBPTR = 0xd
Sep 2 12:01:27 surgeon kernel: SCSISEQ = 0x12, SBLKCTL = 0x0
Sep 2 12:01:27 surgeon kernel: DFCNTRL = 0x78, DFSTATUS = 0x0
Sep 2 12:01:27 surgeon kernel: LASTPHASE = 0x40, SCSISIGI = 0x44, SXFRCTL0 = 0x80
Sep 2 12:01:27 surgeon kernel: SSTAT0 = 0x0, SSTAT1 = 0x2
Sep 2 12:01:27 surgeon kernel: STACK == 0x96, 0x186, 0x156, 0x0
Sep 2 12:01:27 surgeon kernel: SCB count = 60
Sep 2 12:01:27 surgeon kernel: Kernel NEXTQSCB = 46
Sep 2 12:01:27 surgeon kernel: Card NEXTQSCB = 34
Sep 2 12:01:27 surgeon kernel: QINFIFO entries: 34 31 45 9 52 54 50 10 33 28 29 35 32 43 47 40 19 13 6 55 39
Sep 2 12:01:27 surgeon kernel: Waiting Queue entries:
Sep 2 12:01:27 surgeon kernel: Disconnected Queue entries:
Sep 2 12:01:27 surgeon kernel: QOUTFIFO entries:
Sep 2 12:01:27 surgeon kernel: Sequencer Free SCB List: 3 7 1 11 4 0 8 9 15 12 2 14 6 5 10
Sep 2 12:01:27 surgeon kernel: Sequencer SCB Info: 0(c 0x60, s 0x7, l 0, t 0xff) 1(c 0x60, s 0x7, l 0, t 0xff) 2(c 0x60, s 0x7, l 0, t 0xff) 3(c 0x60, s 0x7, l 0, t 0xff) 4(c 0x60, s 0x7, l 0, t 0xff) 5(c 0x60, s 0x7, l 0, t 0xff) 6(c 0x60, s 0x7, l 0, t 0xff) 7(c 0x60, s 0x7, l 0, t 0xff) 8(c 0x60, s 0x7, l 0, t 0xff) 9(c 0x60, s 0x7, l 0, t 0xff) 10(c 0x60, s 0x7, l 0, t 0xff) 11(c 0x60, s 0x7, l 0, t 0xff) 12(c 0x60, s 0x7, l 0, t 0xff) 13(c 0x60, s 0x7, l 0, t 0x29) 14(c 0x60, s 0x7, l 0, t 0xff) 15(c 0x60, s 0x7, l 0, t 0xff)
Sep 2 12:01:27 surgeon kernel: Pending list: 39(c 0x60, s 0x7, l 0), 55(c 0x60, s 0x7, l 0), 6(c 0x60, s 0x7, l 0), 13(c 0x60, s 0x7, l 0), 19(c 0x60, s 0x7, l 0), 40(c 0x60, s 0x7, l 0), 47(c 0x60, s 0x7, l 0), 43(c 0x60, s 0x7, l 0), 32(c 0x60, s 0x7, l 0), 35(c 0x60, s 0x7, l 0), 29(c 0x60, s 0x7, l 0), 28(c 0x60, s 0x7, l 0), 33(c 0x60, s 0x7, l 0), 10(c 0x60, s 0x7, l 0), 50(c 0x60, s 0x7, l 0), 54(c 0x60, s 0x7, l 0), 52(c 0x60, s 0x7, l 0), 9(c 0x60, s 0x7, l 0), 45(c 0x60, s 0x7, l 0), 31(c 0x60, s 0x7, l 0), 34(c 0x60, s 0x7, l 0), 41(c 0x60, s 0x7, l 0)
Sep 2 12:01:27 surgeon kernel: Kernel Free SCB list: 1 44 48 51 36 37 38 49 42 11 27 2 0 16 3 12 25 17 53 59 58 22 23 4 8 24 57 7 30 26 5 21 15 14 20 18 56
Sep 2 12:01:27 surgeon kernel: DevQ(0:0:0): 0 waiting
Sep 2 12:01:27 surgeon kernel: DevQ(0:5:0): 0 waiting
Sep 2 12:01:27 surgeon kernel: DevQ(0:6:0): 0 waiting
Sep 2 12:01:27 surgeon kernel: scsi0:0:0:0: Cmd aborted from QINFIFO
Sep 2 12:01:27 surgeon kernel: aic7xxx_abort returns 0x2002
Sep 2 12:01:27 surgeon kernel: scsi0:0:0:0: Attempting to queue an ABORT message
Sep 2 12:01:27 surgeon kernel: scsi0: Dumping Card State in Data-in phase, at SEQADDR 0x7a
Sep 2 12:01:27 surgeon kernel: ACCUM = 0x0, SINDEX = 0x8, DINDEX = 0x8f, ARG_2 = 0xff
Sep 2 12:01:27 surgeon kernel: HCNT = 0x18 SCBPTR = 0xd
Sep 2 12:01:27 surgeon kernel: SCSISEQ = 0x12, SBLKCTL = 0x0
Sep 2 12:01:27 surgeon kernel: DFCNTRL = 0x78, DFSTATUS = 0x0
Sep 2 12:01:27 surgeon kernel: LASTPHASE = 0x40, SCSISIGI = 0x44, SXFRCTL0 = 0x80
Sep 2 12:01:27 surgeon kernel: SSTAT0 = 0x0, SSTAT1 = 0x2
Sep 2 12:01:27 surgeon kernel: STACK == 0x96, 0x186, 0x156, 0x0
Sep 2 12:01:27 surgeon kernel: SCB count = 60
Sep 2 12:01:28 surgeon kernel: Kernel NEXTQSCB = 34
Sep 2 12:01:28 surgeon kernel: Card NEXTQSCB = 46
Sep 2 12:01:28 surgeon kernel: QINFIFO entries: 46 31 45 9 52 54 50 10 33 28 29 35 32 43 47 40 19 13 6 55
Sep 2 12:01:28 surgeon kernel: Waiting Queue entries:
Sep 2 12:01:28 surgeon kernel: Disconnected Queue entries:
Sep 2 12:01:28 surgeon kernel: QOUTFIFO entries:
Sep 2 12:01:28 surgeon kernel: Sequencer Free SCB List: 3 7 1 11 4 0 8 9 15 12 2 14 6 5 10
Sep 2 12:01:28 surgeon kernel: Sequencer SCB Info: 0(c 0x60, s 0x7, l 0, t 0xff) 1(c 0x60, s 0x7, l 0, t 0xff) 2(c 0x60, s 0x7, l 0, t 0xff) 3(c 0x60, s 0x7, l 0, t 0xff) 4(c 0x60, s 0x7, l 0, t 0xff) 5(c 0x60, s 0x7, l 0, t 0xff) 6(c 0x60, s 0x7, l 0, t 0xff) 7(c 0x60, s 0x7, l 0, t 0xff) 8(c 0x60, s 0x7, l 0, t 0xff) 9(c 0x60, s 0x7, l 0, t 0xff) 10(c 0x60, s 0x7, l 0, t 0xff) 11(c 0x60, s 0x7, l 0, t 0xff) 12(c 0x60, s 0x7, l 0, t 0xff) 13(c 0x60, s 0x7, l 0, t 0x29) 14(c 0x60, s 0x7, l 0, t 0xff) 15(c 0x60, s 0x7, l 0, t 0xff)
Sep 2 12:01:28 surgeon kernel: Pending list: 55(c 0x60, s 0x7, l 0), 6(c 0x60, s 0x7, l 0), 13(c 0x60, s 0x7, l 0), 19(c 0x60, s 0x7, l 0), 40(c 0x60, s 0x7, l 0), 47(c 0x60, s 0x7, l 0), 43(c 0x60, s 0x7, l 0), 32(c 0x60, s 0x7, l 0), 35(c 0x60, s 0x7, l 0), 29(c 0x60, s 0x7, l 0), 28(c 0x60, s 0x7, l 0), 33(c 0x60, s 0x7, l 0), 10(c 0x60, s 0x7, l 0), 50(c 0x60, s 0x7, l 0), 54(c 0x60, s 0x7, l 0), 52(c 0x60, s 0x7, l 0), 9(c 0x60, s 0x7, l 0), 45(c 0x60, s 0x7, l 0), 31(c 0x60, s 0x7, l 0), 46(c 0x60, s 0x7, l 0), 41(c 0x60, s 0x7, l 0)
Sep 2 12:01:28 surgeon kernel: Kernel Free SCB list: 39 1 44 48 51 36 37 38 49 42 11 27 2 0 16 3 12 25 17 53 59 58 22 23 4 8 24 57 7 30 26 5 21 15 14 20 18 56
Sep 2 12:01:28 surgeon kernel: DevQ(0:0:0): 0 waiting
Sep 2 12:01:28 surgeon kernel: DevQ(0:5:0): 0 waiting
Sep 2 12:01:28 surgeon kernel: DevQ(0:6:0): 0 waiting
Sep 2 12:01:28 surgeon kernel: scsi0:0:0:0: Cmd aborted from QINFIFO
Sep 2 12:01:28 surgeon kernel: aic7xxx_abort returns 0x2002
....
... (this cycle also repeats a few dozen times)
Hi,
On Tue, 3 Sep 2002, Jeffrey J. Kosowsky wrote:
> I am running a pretty standard RH7.3 configuration (Kernel 2.4.18)
> with root and key filesystems on a 20Gig EIDE disk.
I think we fixed that for 2.4.19.
Thunder
--
--./../...-/. -.--/---/..-/.-./..././.-../..-. .---/..-/.../- .-
--/../-./..-/-/./--..-- ../.----./.-../.-.. --./../...-/. -.--/---/..-
.- -/---/--/---/.-./.-./---/.--/.-.-.-
--./.-/-.../.-./.././.-../.-.-.-
On Tue, 2002-09-03 at 16:23, Thunder from the hill wrote:
> Hi,
>
> On Tue, 3 Sep 2002, Jeffrey J. Kosowsky wrote:
> > I am running a pretty standard RH7.3 configuration (Kernel 2.4.18)
> > with root and key filesystems on a 20Gig EIDE disk.
>
> I think we fixed that for 2.4.19.
I don't
The problem he is reporting is also on the SCSI side. Now two things
have changed over time
#1 The 2.4 scsi driver (Justin's notably so) is more aggressive than 2.0
ever was about speeds. It will pick Ultra80 or Ultra160 if the bus
appears capable and the drives do. The old 2.0 code probably never
picked anything more enterprising than 20Mhz.
That does mean that marginal termination will show up when it was
apparently ok both in Linux and in older windows which was also
generally rather conservative
#2 Its a newer driver. aic7xxx_old is the driver we used to use and
trying that might be informative too.