Gentoo Archives: gentoo-user

From: Richard Broersma Jr <rabroersma@×××××.com>
To: gentoo-user@l.g.o
Subject: Re: [gentoo-user] dma_intr errors on heavy writes -- cause for concern?
Date: Wed, 01 Nov 2006 06:18:21
Message-Id: 20061101061416.44840.qmail@web31814.mail.mud.yahoo.com
In Reply to: Re: [gentoo-user] dma_intr errors on heavy writes -- cause for concern? by Richard Fish
1 > On 10/31/06, Richard Broersma Jr <rabroersma@×××××.com> wrote:
2 > > During certain times I my server under goes heavy disc writes. When this happen I get the
3 > > following error. Should I be concerned? Does anyone know of any resources that I can read up
4 > on
5 > > that will explain that all of this mean?
6 >
7 > I don't have any resources, but my understanding is that LBAsect is
8 > the logical sector of the block device (i.e, the raid array), while
9 > sector is the physical sector of disk. LBAsect looks
10 > suspicious...referencing a sector that is somewhere around the 91
11 > petabytes address. You didn't mention how large the disks are, but
12 > even there they are requesting a sector that is about 220G from the
13 > beginning of the disk, and returning that no such sector exists
14 > (SectorIdNotFound).
15 >
16 > So my guess is that your filesystem is getting confused under load,
17 > and trying to access stuff that is beyond the end of your raid array.
18 > So, which fs and kernel version?
19
20 oops, I was mistake, I forgot that when I re-arrange my disks my RAID10 is partly using hda/hdc.
21
22 Linux version 2.6.17-gentoo-r7 (root@db_server01)
23 (gcc version 4.1.1 (Gentoo 4.1.1)) #8 Sun Oct 8 20:28:34 PDT 2006
24
25 md4 : active raid10 hdg1[3] hde1[2] hdc1[1] hda1[0]
26 586098688 blocks 1024K chunks 2 near-copies [4/4] [UUUU]
27 fstab
28 /dev/md4 /home ext3 noatime 0 2
29 df
30 /dev/md/4 576901664 7284500 540312232 2% /home
31
32 Disk /dev/hda: 320.0 GB, 320072933376 bytes
33 255 heads, 63 sectors/track, 38913 cylinders
34 Units = cylinders of 16065 * 512 = 8225280 bytes
35 Device Boot Start End Blocks Id System
36 /dev/hda1 1 38913 312568641 fd Linux raid autodetect
37
38 Disk /dev/hdc: 320.0 GB, 320072933376 bytes
39 255 heads, 63 sectors/track, 38913 cylinders
40 Units = cylinders of 16065 * 512 = 8225280 bytes
41 Device Boot Start End Blocks Id System
42 /dev/hdc1 1 38913 312568641 fd Linux raid autodetect
43
44 Disk /dev/hde: 320.0 GB, 320072933376 bytes
45 255 heads, 63 sectors/track, 38913 cylinders
46 Units = cylinders of 16065 * 512 = 8225280 bytes
47 Device Boot Start End Blocks Id System
48 /dev/hde1 1 38913 312568641 fd Linux raid autodetect
49
50 Disk /dev/hdg: 300.0 GB, 300090728448 bytes
51 255 heads, 63 sectors/track, 36483 cylinders
52 Units = cylinders of 16065 * 512 = 8225280 bytes
53 Device Boot Start End Blocks Id System
54 /dev/hdg1 1 36483 293049666 fd Linux raid autodetect
55
56 Regards,
57
58 Richard Broersma Jr.
59
60 --
61 gentoo-user@g.o mailing list

Replies

Subject Author
Re: [gentoo-user] dma_intr errors on heavy writes -- cause for concern? Richard Fish <bigfish@××××××××××.org>