SCSI Errors on Oracle Server

[Stephen Carville]

One of my Oracle servers is generating SCSI errors I cannot resolve.  I've 
googled the errors and found lots of questions but no answers.  The only 
common factor I've noticed is all the compliants seem to involve machines 
with SMP and software RAID on Adaptec controllers or tape drives.   

Anyways, does anyone know what causes these errors and how I can cure it?

Summary of configuration

Make and Model:		Dell PowerEdge 2650
Processors:	 		2 x 3.2Ghz Xeon Processors w/hyperthreading enabled
OS: 					Redhat AS 3.0
Kernel Version:			2.4.21-4.ELsmp
Main Software:			Oracle 9i
Boot and System Drives	2 X 146G SCSI Drive in software RAID 1.
Data Drives:			6 x 146G SCSI drives in software RAID 10 Array

Example errors

Jan 14 09:33:26 tigris kernel: scsi0: Unexpected busfree while idle
Jan 14 09:33:26 tigris kernel: SEQADDR == 0x29
Jan 14 09:34:26 tigris kernel: scsi0:0:6:0: Attempting to queue an
ABORT message
Jan 14 09:34:26 tigris kernel: CDB: 0x12 0x0 0x0 0x0 0xff 0x0
Jan 14 09:34:26 tigris kernel: scsi0: At time of recovery, card was
not paused
Jan 14 09:34:26 tigris kernel: >>>>>>>>>>>>>>>>>> Dump Card State
Begins <<<<<<<
<<<<<<<<<<
Jan 14 09:34:26 tigris kernel: scsi0: Dumping Card State while idle,
at SEQADDR 0x8
Jan 14 09:34:26 tigris kernel: Card was paused
Jan 14 09:34:26 tigris kernel: ACCUM = 0x0, SINDEX = 0x73, DINDEX =
0xe4, ARG_2 = 0x0
Jan 14 09:34:26 tigris kernel: HCNT = 0x0 SCBPTR = 0x8
Jan 14 09:34:26 tigris kernel: SCSIPHASE[0x0] SCSISIGI[0x0] ERROR[0x0]
SCSIBUSL[0x0]
Jan 14 09:34:26 tigris kernel: LASTPHASE[0x1] SCSISEQ[0x12]
SBLKCTL[0xa] SCSIRATE[0x0]
Jan 14 09:34:26 tigris kernel: SEQCTL[0x10] SEQ_FLAGS[0xc0]
SSTAT0[0x0] SSTAT1[0x8]
Jan 14 09:34:26 tigris kernel: SSTAT2[0x0] SSTAT3[0x0] SIMODE0[0x8]
SIMODE1[0xa4]
Jan 14 09:34:26 tigris kernel: SXFRCTL0[0x80] DFCNTRL[0x0] DFSTATUS[0x89]
Jan 14 09:34:26 tigris kernel: STACK: 0x2f 0x16a 0x110 0x3
Jan 14 09:34:26 tigris kernel: SCB count = 198
Jan 14 09:34:26 tigris kernel: Kernel NEXTQSCB = 117
Jan 14 09:34:26 tigris kernel: Card NEXTQSCB = 117
Jan 14 09:34:26 tigris kernel: QINFIFO entries:
Jan 14 09:34:26 tigris kernel: Waiting Queue entries:
Jan 14 09:34:26 tigris kernel: Disconnected Queue entries:
Jan 14 09:34:26 tigris kernel: QOUTFIFO entries:
Jan 14 09:34:26 tigris kernel: Sequencer Free SCB List: 8 17 27 22 14
10 25 20 31 1 4 15 30 18 23 28 26 11 3 2 16 12 0 6 29 9 24 7 19 13 5
Jan 14 09:34:26 tigris kernel: Sequencer SCB Info:
Jan 14 09:34:26 tigris kernel: 0 SCB_CONTROL[0xe0] SCB_SCSIID[0x17]
SCB_LUN[0x0] SCB_TAG[0xff]
Jan 14 09:34:26 tigris kernel: 1 SCB_CONTROL[0xe0] SCB_SCSIID[0x17]
SCB_LUN[0x0] SCB_TAG[0xff]
Jan 14 09:34:26 tigris kernel: 2 SCB_CONTROL[0xe0] SCB_SCSIID[0x57]
SCB_LUN[0x0 SCB_TA
G[0xff]
Jan 14 09:34:27 tigris kernel: Pending list:
Jan 14 09:34:27 tigris kernel: 98 SCB_CONTROL[0x40] SCB_SCSIID[0x67]
SCB_LUN[0x0]
Jan 14 09:34:27 tigris kernel: Kernel Free SCB list: 115 25 132 135
151 93 13 21 64 147 142 106 140 1
58 54 105 0 122 50 30 80 56 120 65 177 131 61 128 87 24 1 148 134 43 8
88 38 35
129 139 45 165 168 190 44 162 17 68 152 103 164 34 85 41 6 174 145 101
180 138 1
27 176 175 179 118 183 84 167 94 116 58 189 7 74 22 137 114 161 155 49
149 188 7
5 90 12 77 107 10 20 156 95 108 51 96 52 28 102 169 18 92 163 69 133
57 112 154
170 99 55 11 66 37 172 186 191 119 144 63 83 81 33 3 153 97 100 36 121
5 48 91 1
30 89 23 171 109 124 2 72 197 86 42 181 19 16 113 173 111 62 178 73 4
182 126 19
6 166 76 157 39 79 82 143 53 141 159 184 123 26 125 187 47 59 146 104
29 78 185
110 71 46 60 27 9 14 32 67 136 70 160 40 15 31 150 195 194 193 192
Jan 14 09:34:27 tigris kernel: Untagged Q(6): 98
Jan 14 09:34:27 tigris kernel: DevQ(0:0:0): 0 waiting
Jan 14 09:34:27 tigris kernel: DevQ(0:1:0): 0 waiting
Jan 14 09:34:27 tigris kernel: DevQ(0:2:0): 0 waiting
Jan 14 09:34:27 tigris kernel: DevQ(0:3:0): 0 waiting
Jan 14 09:34:27 tigris kernel: DevQ(0:4:0): 0 waiting
Jan 14 09:34:27 tigris kernel: DevQ(0:5:0): 0 waiting
Jan 14 09:34:27 tigris kernel: DevQ(0:6:0): 0 waiting
Jan 14 09:34:27 tigris kernel: DevQ(0:8:0): 0 waiting
Jan 14 09:34:27 tigris kernel:
Jan 14 09:34:27 tigris kernel: <<<<<<<<<<<<<<<<< Dump Card State Ends
Jan 14 09:34:27 tigris kernel: (scsi0:A:6:0): Device is disconnected, 
re-queuing SCB
Jan 14 09:34:27 tigris kernel: (scsi0:A:6:0): Recovery code sleeping
Jan 14 09:34:27 tigris kernel: Abort Message Sent
Jan 14 09:34:27 tigris kernel: (scsi0:A:6:0): SCB 98 - Abort Completed.
Jan 14 09:34:27 tigris kernel: Recovery SCB completes
Jan 14 09:34:27 tigris kernel: Recovery code awake
Jan 14 09:34:27 tigris kernel: aic7xxx_abort returns 0x2002

-- 
Stephen Carville
Unix and Network Adminstrator
Nationwide-Totalflood
6033 W.Century Blvd.
Los Angeles, CA 90045
310-342-3602