[BlueOnyx:13387] attempt to access beyond end of device

[BO Mail]

Hi,
Wondering if anyone has seen this error and/or knows if/how its fixable? This
just started a couple weeks ago and sometimes it will go a day without even
seeing this error message in /var/messages. Other times we will see it off and
on 4 or 5 times a day.


attempt to access beyond end of device
md4: rw=0, want=8942984760, limit=287193728

saw this message once and only once in the past day:

kernel: md4: rw=1, want=3302887440, limit=287193728
kernel: Buffer I/O error on device md4, logical block 412860929
kernel: lost page write due to I/O error on md4


Other than these messages everything seems fine on the server, no other error
messages and everything running normally. It has a raid1 mirror with 2 160gb
seagate HD's and has been running along smoothly for a few years. We did
smartctl -a checks and they say they passed. Did have to replace the original
sdb drive 2 or 3 yrs ago as it failed. 


cat /proc/mdstat
Personalities : [raid1]
md6 : active raid1 sdb1[0] sda1[1]
      104320 blocks [2/2] [UU]

md3 : active raid1 sdb3[0] sda3[1]
      4192896 blocks [2/2] [UU]

md5 : active raid1 sdb5[0] sda5[1]
      1052160 blocks [2/2] [UU]

md2 : active raid1 sdb6[0] sda6[1]
      1052160 blocks [2/2] [UU]

md4 : active raid1 sdb7[0] sda7[1]
      143596864 blocks [2/2] [UU]

md1 : active raid1 sdb2[0] sda2[1]
      6289344 blocks [2/2] [UU]

unused devices: <none>

kernel: md: md driver 0.90.0 MAX_MD_DEVS=256, MD_SB_DISKS=27
kernel: ata1: SATA max UDMA/133 cmd 0xE200 ctl 0xE302 bmdma 0xE600 irq 177
kernel: ata2: SATA max UDMA/133 cmd 0xE400 ctl 0xE502 bmdma 0xE608 irq 177
kernel: md: raid1 personality registered as nr 3
kernel: md: Autodetecting RAID arrays.
kernel: md: autorun ...
kernel: md: considering sdb7 ...
kernel: md:  adding sdb7 ...
kernel: md: sdb6 has different UUID to sdb7
kernel: md: sdb5 has different UUID to sdb7
kernel: md: sdb3 has different UUID to sdb7
kernel: md: sdb2 has different UUID to sdb7
kernel: md: sdb1 has different UUID to sdb7
kernel: md:  adding sda7 ...
kernel: md: sda6 has different UUID to sdb7
kernel: md: sda5 has different UUID to sdb7
kernel: md: sda3 has different UUID to sdb7
kernel: md: sda2 has different UUID to sdb7
kernel: md: sda1 has different UUID to sdb7
kernel: md: created md4
kernel: md: bind<sda7>
kernel: md: bind<sdb7>
kernel: md: running: <sdb7><sda7>
kernel: raid1: raid set md4 active with 2 out of 2 mirrors
kernel: md: considering sdb6 ...
kernel: md:  adding sdb6 ...
kernel: md: sdb5 has different UUID to sdb6
kernel: md: sdb3 has different UUID to sdb6
kernel: md: sdb2 has different UUID to sdb6
kernel: md: sdb1 has different UUID to sdb6
kernel: md:  adding sda6 ...
kernel: md: sda5 has different UUID to sdb6
kernel: md: sda3 has different UUID to sdb6
vwh kernel: md: sda2 has different UUID to sdb6
kernel: md: sda1 has different UUID to sdb6
kernel: md: created md2
kernel: md: bind<sda6>
kernel: md: bind<sdb6>
kernel: md: running: <sdb6><sda6>
kernel: raid1: raid set md2 active with 2 out of 2 mirrors
kernel: md: considering sdb5 ...
kernel: md:  adding sdb5 ...
kernel: md: sdb3 has different UUID to sdb5
kernel: md: sdb2 has different UUID to sdb5
kernel: md: sdb1 has different UUID to sdb5
kernel: md:  adding sda5 ...
kernel: md: sda3 has different UUID to sdb5
kernel: md: sda2 has different UUID to sdb5
kernel: md: sda1 has different UUID to sdb5
kernel: md: created md5
kernel: md: bind<sda5>
kernel: md: bind<sdb5>
kernel: md: running: <sdb5><sda5>
kernel: raid1: raid set md5 active with 2 out of 2 mirrors
kernel: md: considering sdb3 ...
kernel: md:  adding sdb3 ...
kernel: md: sdb2 has different UUID to sdb3
kernel: md: sdb1 has different UUID to sdb3
kernel: md:  adding sda3 ...
kernel: md: sda2 has different UUID to sdb3
kernel: md: sda1 has different UUID to sdb3
kernel: md: created md3
kernel: md: bind<sda3>
kernel: md: bind<sdb3>
kernel: md: running: <sdb3><sda3>
kernel: raid1: raid set md3 active with 2 out of 2 mirrors
kernel: md: considering sdb2 ...
kernel: md:  adding sdb2 ...
kernel: md: sdb1 has different UUID to sdb2
kernel: md:  adding sda2 ...
kernel: md: sda1 has different UUID to sdb2
kernel: md: created md1
kernel: md: bind<sda2>
kernel: md: bind<sdb2>
kernel: md: running: <sdb2><sda2>
kernel: raid1: raid set md1 active with 2 out of 2 mirrors
kernel: md: considering sdb1 ...
kernel: md:  adding sdb1 ...
kernel: md:  adding sda1 ...
kernel: md: created md6
kernel: md: bind<sda1>
kernel: md: bind<sdb1>
kernel: md: running: <sdb1><sda1>
kernel: raid1: raid set md6 active with 2 out of 2 mirrors
kernel: md: ... autorun DONE.
kernel: md: Autodetecting RAID arrays.
kernel: md: autorun ...
kernel: md: ... autorun DONE.
kernel: md: Autodetecting RAID arrays.
kernel: md: autorun ...
kernel: md: ... autorun DONE.
kernel: md: Autodetecting RAID arrays.
kernel: md: autorun ...
kernel: md: ... autorun DONE.
kernel: md: Autodetecting RAID arrays.
kernel: md: autorun ...
kernel: md: ... autorun DONE.
kernel: md: Autodetecting RAID arrays.
kernel: md: autorun ...
kernel: md: ... autorun DONE.
kernel: md: Autodetecting RAID arrays.
kernel: md: autorun ...
kernel: md: ... autorun DONE.
kernel: EXT3 FS on md1, internal journal
kernel: EXT3 FS on md6, internal journal
kernel: EXT3 FS on md4, internal journal
kernel: EXT3 FS on md2, internal journal
kernel: EXT3 FS on md3, internal journal
kernel: Adding 1052152k swap on /dev/md5.  Priority:-1 extents:1

Filesystem           1K-blocks      Used Available Use% Mounted on
/dev/md1               6190592    940700   4935428  17% /
/dev/md6                101018     15385     80417  17% /boot
none                    509160         0    509160   0% /dev/shm
/dev/md4             141342888   2106040 132057008   2% /home
/dev/md2               1035596     78316    904672   8% /tmp
/dev/md3               4127040    490928   3426468  13% /var


Oh and this is a BQ server, our last server holdout from switching to
BlueOnyx. They say they have some "custom" stuff that won't work yet on newer
OS, but they are working on it so we can switch over to BO eventually. If
anyone has any ideas we would greatly appreciate them.


Thanks,
Tom