amr intermittant failures

[Len Conrad]

uname -a
FreeBSD mx6 7.2-RELEASE FreeBSD 7.2-RELEASE #0: Fri May  1 07:18:07 UTC 2009     root at driscoll.cse.buffalo.edu:/usr/obj/usr/src/sys/GENERIC  amd64

dmesg.boot:

amr0: <LSILogic MegaRAID 1.53> mem 0xf80f0000-0xf80fffff,0xfe9e0000-0xfe9fffff irq 46 at device 14.0 on pci2
amr0: Using 64-bit DMA
amr0: [ITHREAD]
amr0: delete logical drives supported by controller
amr0: <LSILogic PERC 4e/Si> Firmware 522A, BIOS H430, 256MB RAM

errors:

Sep 21 18:01:32 mx6 kernel: amr0: Too many retries on command 0xfffffffe80229fc8.  Controller is likely dead
Sep 21 21:51:32 mx6 kernel: amr0: Too many retries on command 0xfffffffe80229970.  Controller is likely dead
Sep 22 00:36:31 mx6 kernel: amr0: Too many retries on command 0xfffffffe802201e0.  Controller is likely dead
Sep 22 03:01:36 mx6 kernel: amr0: Too many retries on command 0xfffffffe80229b40.  Controller is likely dead
Sep 22 03:02:25 mx6 kernel: amr0: Too many retries on command 0xfffffffe80228af0.  Controller is likely dead
Sep 22 03:05:54 mx6 kernel: amr0: Too many retries on command 0xfffffffe80228838.  Controller is likely dead
Sep 22 09:36:29 mx6 kernel: amr0: Too many retries on command 0xfffffffe80226d60.  Controller is likely dead
Sep 22 09:41:29 mx6 kernel: amr0: Too many retries on command 0xfffffffe8022a7f0.  Controller is likely dead
Sep 22 10:11:29 mx6 kernel: amr0: Too many retries on command 0xfffffffe80229ee0.  Controller is likely dead
Sep 22 10:56:29 mx6 kernel: amr0: Too many retries on command 0xfffffffe80229b40.  Controller is likely dead
Sep 22 11:16:29 mx6 kernel: amr0: Too many retries on command 0xfffffffe80224668.  Controller is likely dead
Sep 22 11:31:29 mx6 kernel: amr0: Too many retries on command 0xfffffffe80224f78.  Controller is likely dead
Sep 22 16:46:28 mx6 kernel: amr0: Too many retries on command 0xfffffffe80226c78.  Controller is likely dead
Sep 22 16:51:28 mx6 kernel: amr0: Too many retries on command 0xfffffffe80229230.  Controller is likely dead
Sep 22 22:01:27 mx6 kernel: amr0: Too many retries on command 0xfffffffe80227758.  Controller is likely dead
Sep 23 03:02:38 mx6 kernel: amr0: Too many retries on command 0xfffffffe80225400.  Controller is likely dead
Sep 23 03:03:42 mx6 kernel: amr0: Too many retries on command 0xfffffffe802260b0.  Controller is likely dead
Sep 23 03:06:05 mx6 kernel: amr0: Too many retries on command 0xfffffffe80225970.  Controller is likely dead
Sep 23 03:11:26 mx6 kernel: amr0: Too many retries on command 0xfffffffe802280f8.  Controller is likely dead
Sep 23 03:16:26 mx6 kernel: amr0: Too many retries on command 0xfffffffe80226aa8.  Controller is likely dead
Sep 23 04:41:26 mx6 kernel: amr0: Too many retries on command 0xfffffffe802294e8.  Controller is likely dead
Sep 23 09:21:25 mx6 kernel: amr0: Too many retries on command 0xfffffffe80225318.  Controller is likely dead
Sep 23 09:41:25 mx6 kernel: amr0: Too many retries on command 0xfffffffe80228a08.  Controller is likely dea
Sep 23 12:01:25 mx6 kernel: amr0: Too many retries on command 0xfffffffe802296b8.  Controller is likely dead
Sep 23 12:06:25 mx6 kernel: amr0: Too many retries on command 0xfffffffe802241e0.  Controller is likely dead
Sep 23 12:36:24 mx6 kernel: amr0: Too many retries on command 0xfffffffe802281e0.  Contro
Sep 23 14:06:24 mx6 kernel: amr0: Too many retries on command 0xfffffffe80228bd8.  Controller is likely dead
Sep 23 14:21:24 mx6 kernel: amr0: Too many retries on command 0xfffffffe80225230.  Controller is likely dead

these are apparently non fatal.  VERY busy relay-only postfix MX boxes, two of them have the same errors.  Disk is getting hit hard with postfix logging and msg queueing.

Len