Interesting Linux boot up log

[dogwomble]

Hey guys,

I'm progressively working on my Linux home server and ever since I've built the box, I've had this weird problem.

To make things a bit easier, this is how I've got my drives laid out:

HDA: 13gb "System" Drive (ie. holds OS)
HDB: 80gb "Data" drive (holds all my settings, work files etc.)
HDC: DVD Rom Drive
HDD: Nothing
HDE: 320gb drive for RAID
HDF: 320gb drive for RAID
HDG: 320gb drive for RAID
HDH: 320gb drive for RAID

The 320's are on a separate PCI RAID card.

When I boot up the system, for a period of about 30 seconds I get the following message repeated:

Aug 1 23:20:25 kelpie kernel: hde: task_in_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
Aug 1 23:20:25 kelpie kernel: hde: task_in_intr: error=0x10 { SectorIdNotFound }, LBAsect=2500549184, high=149, low=744000, sector=2500549184
Aug 1 23:20:25 kelpie kernel: ide: failed opcode was: unknown

Now being a Linux 'novice', I expect this is bad. However, the system still boots and I can access the RAID partition OK. The only thing I can think of is that, because this happens so early in the boot sequence, then it's possibly just caused by the drives not fully spun up yet or something similar.

If somebody could shed some light onto the matter before this thing goes into production, that would be excellent :)

[Asmodeus]

could it be looping and waiting for the drives to spin up, sync, initialize, and so on?

why do I want ot say most raid cards read as scsi, not ide..

been a while for me with errors like this. and are you doing raid through the card or making a soft said through kernel services... and does going the other route make a difference?

[dogwomble]

At this stage, it's doing RAID in hardware. I haven't tried in software purely because breaking the hardware RAID, then reassembling it as software RAID, is something I'm not quite sure I'm up to yet...

[sagit]

perhaps its because the disks are so large. maybe you need to set something to indicate this (like the windoze reg setting "enable lba large" (or whatever it is)

[Spingo]

Well, if you have RAID set up in Hardware, I don't understand why the RAID controller is passing on four seperate 320GB drives to the OS instead of one big ~960GB drive (assuming you're using RAID-5).

Methinks that your kernel does not support your RAID card and does not have a suitable module available to detect the hardware RAID. Instead of seeing you're RAID device, it's then detecting four seperate drives.

What make/model of RAID card are you using, what flavour of Linux are you using, and what kernel packages have you got installed?

[dogwomble]

OK, the RAID card I'm using is a dual-port (ie. two IDE channel) version of this (http://www.peridyme.com.au/index.php?fuseaction=product.view&productid=129), which from the website appears to be based on the Silicon 680 chipeset. And probably much to your disgust, it's running Fedora Core 5 - puely because it's what I have :) (After trying to find a number of 'decent' distros, I got sick of downloading new ones....)

And FYI here is a list of modules currently compiled into the kernel (Spingo if you can safely guide me through getting rid of any that would not really be needed then you'd be a champion and a scholar ... but I know you're a busy man and I do appreciate your help....)

[root@kelpie /]# lsmod
Module Size Used by
xt_pkttype 1985 0
iptable_raw 2241 0
xt_CLASSIFY 2369 0
xt_CONNMARK 2497 0
xt_length 2113 0
xt_connmark 2113 0
xt_physdev 2385 0
xt_conntrack 2625 0
ipt_ULOG 7749 0
ipt_TTL 2497 0
ipt_ttl 1985 0
ipt_TOS 2497 0
ipt_tos 1729 0
ipt_TCPMSS 4161 0
ipt_SAME 2497 0
ipt_REDIRECT 2241 0
ipt_recent 10189 0
ipt_policy 3009 0
ipt_owner 2241 0
ipt_NETMAP 2241 0
ipt_multiport 2497 0
ipt_MASQUERADE 3777 0
ipt_iprange 1857 0
ipt_hashlimit 9161 0
ipt_esp 1985 0
ipt_ECN 3265 0
ipt_ecn 2241 0
ipt_DSCP 2497 0
ipt_dscp 1857 0
ipt_CLUSTERIP 8261 0
ipt_ah 1985 0
ipt_addrtype 2113 0
ip_nat_irc 2753 0
ip_nat_tftp 1985 0
ip_nat_ftp 3393 0
ip_conntrack_irc 6833 1 ip_nat_irc
ip_conntrack_tftp 4281 1 ip_nat_tftp
ip_conntrack_ftp 7601 1 ip_nat_ftp
iptable_nat 8133 0
ip_nat 16621 8 ipt_SAME,ipt_REDIRECT,ipt_NETMAP,ipt_MASQUERADE, ip_nat_irc,ip_nat_tftp,ip_nat_ftp,iptable_nat
ip_conntrack 49261 12 xt_CONNMARK,xt_connmark,xt_conntrack,ipt_MASQUE RADE,ip_nat_irc,ip_nat_tftp,ip_nat_ftp,ip_conntrac k_irc,ip_conntrack_tftp,ip_con ntrack_ftp,iptable_nat,ip_nat
nfnetlink 6489 2 ip_nat,ip_conntrack
iptable_mangle 2881 0
iptable_filter 3137 0
ip_tables 11657 4 iptable_raw,iptable_nat,iptable_mangle,iptable_f ilter
ipv6 225569 10
xt_mark 1985 0
xt_MARK 2625 0
ipt_REJECT 5441 0
xt_tcpudp 3265 0
ipt_LOG 6081 0
x_tables 12613 38 xt_pkttype,xt_CLASSIFY,xt_CONNMARK,xt_length,xt _connmark,xt_physdev,xt_conntrack,ipt_ULOG,ipt_TTL ,ipt_ttl,ipt_TOS,ipt_tos,ipt_T CPMSS,ipt_SAME,ipt_REDIRECT,ipt_recent,ipt_policy, ipt_owner,ipt_NETMAP,ipt_multi port,ipt_MASQUERADE,ipt_iprange,ipt_hashlimit,ipt_ esp,ipt_ECN,ipt_ecn,ipt_DSCP,i pt_dscp,ipt_CLUSTERIP,ipt_ah,ipt_addrtype,iptable_ nat,ip_tables,xt_mark,xt_MARK, ipt_REJECT,xt_tcpudp,ipt_LOG
video 14917 0
button 6609 0
battery 9285 0
ac 4933 0
lp 12297 0
parport_pc 25445 1
parport 34313 2 lp,parport_pc
floppy 57733 0
nvram 8393 0
ohci_hcd 19805 0
ehci_hcd 29005 0
snd_intel8x0 30301 0
snd_ac97_codec 83937 1 snd_intel8x0
snd_ac97_bus 2497 1 snd_ac97_codec
snd_seq_dummy 3781 0
snd_seq_oss 28993 0
snd_seq_midi_event 7105 1 snd_seq_oss
snd_seq 47153 5 snd_seq_dummy,snd_seq_oss,snd_seq_midi_event
i2c_sis96x 5573 0
snd_seq_device 8909 3 snd_seq_dummy,snd_seq_oss,snd_seq
i2c_core 20673 1 i2c_sis96x
snd_pcm_oss 45009 0
snd_mixer_oss 16449 1 snd_pcm_oss
sis900 21953 0
mii 5313 1 sis900
snd_pcm 76869 3 snd_intel8x0,snd_ac97_codec,snd_pcm_oss
r8169 27337 0
snd_timer 22597 2 snd_seq,snd_pcm
snd 50501 9 snd_intel8x0,snd_ac97_codec,snd_seq_oss,snd_seq, snd_seq_device,snd_pcm_oss,snd_mixer_oss,snd_pcm,s nd_timer
soundcore 9377 1 snd
snd_page_alloc 10441 2 snd_intel8x0,snd_pcm
dm_snapshot 16237 0
dm_zero 2241 0
dm_mirror 19985 0
dm_mod 50521 7 dm_snapshot,dm_zero,dm_mirror
ext3 116809 4
jbd 53077 1 ext3
[root@kelpie /]#

[Vally]

Vally is logged on and i can't be bothered logging on as Rusky85 - I know nothing about linux but i have watched them be built @work.

We had a few HDD's in a raid config and it would show up all the HDD's as individual HDD's rather than the 1 single logical drive.

We ended up having to purchase a new raid card for it to work properly.

Don't know if it is a help but i saw above how someone else mentioned that they are suprised it hasn't picked up your logical driver versus the current picking up of each single drive.

[dogwomble]

Vally is logged on and i can't be bothered logging on as Rusky85 - I know nothing about linux but i have watched them be built @work.

We had a few HDD's in a raid config and it would show up all the HDD's as individual HDD's rather than the 1 single logical drive.

We ended up having to purchase a new raid card for it to work properly.

Well, in this case, it's particularly weird.

I'm getting hde/f/g/h, but when I partitioned it it allowed me to do it all as one volume, so I didn't really worry about it becasue that was what I was after. And while I haven't used the card in a while, in previous incarnations of the server, the card seems to have worked pretty well (although I wasn't doing any RAID with it, I was just using it so I could attach more than 4 drives).

[glarbnaught]

Well, in this case, it's particularly weird.

I'm getting hde/f/g/h, but when I partitioned it it allowed me to do it all as one volume, so I didn't really worry about it becasue that was what I was after. And while I haven't used the card in a while, in previous incarnations of the server, the card seems to have worked pretty well (although I wasn't doing any RAID with it, I was just using it so I could attach more than 4 drives).

Being the one that Vally, ehm... sorry Rusky85 watched build the linux box the problem you are having is not related to the same problems I had. I was using a 'fake' hardware RAID controller and well, they don't work so well... any hoo...

See if you can muck around with the hdparm settings on that hard drive or all your hradrives and see if you can't get them all set to the same settings (incase they aren't). Just incase that particular hard drive is being worked a little harder then the others (I am not exactly sure if you can do it on a per drive bases on RAID) otherwise I would say that there is a problem with that particular hard drive and it could be on it's last legs.

[dogwomble]

OK, considering I'm a relatively N00b Linux user, what would I need to modify and what am I looking for?

[exekewtable]

you don't have a real raid card. that chipset doesn't do its raid in hardware, but does it partially in the os. linux doesn't support that sort of shit because the software raid driver is better. You are better off making it a software raid device with that card.
And I would say that that drive is dying. ill jump on irc #zgeek if you are around and walk you through it if you like

[dogwomble]

exeketable,

As you mentioned on IRC last night, I'm beginning to think that it is a hardware problem as you said. I tried the theory of 'overload on the power supply' by removing a few things, but it really didn't change much at all. However, changing the cables seems to have fixed the problem.

However, another problem seems to have occurred. This morning before I went to work I shutdown thesystem using the comand 'init 0'. Now I've got the problem that when it starts up, it just sits there looking pretty at me with tle line "Loading dm-snapshot.ko modules". I've left the machine sitting there for about 15 minutes waiting for it to do something ... with no joy whatsoever.

Ideas?