Problems on Linux power-up

#1

I've had this intermittent problem. Couple of times in perhaps last two
weeks.

I'm running d3 on Red Hat Linux. Not sure of the version numbers, but if
this makes a difference, I can post them.

Switch computer on. Goes through memory & SCSI card check OK.

At some point (and I am not quite certain where as I have not been in from
of the screen) stops with the message:

Loading Linux
Uncompressing Linux
CRC error

{then I think it says - I did not write the last line down)

System stopped ....

Pushing the reboot button has got it starting normally.

My questions are:

Is it serious - likely to get worse?

What can a total ignoramus do to sort it - or do I need an expert?
Bearing on mind that my knowledge of Linux is 0.5 on scale of 1 to 10.
Maybe not even that much!

TIA for any help.

Alan Pritchard

#2

Could be a failing disk. The CRC error is from the boot image being
uncompressed, which is stored in the master boot record.

run "cat /proc/version" and post your verison info.

I would run e2fsck on the disk manually to check for bad blocks,
especially block zero.

Glen

"Alan Pritchard" <alan.pritchard (AT) gmail (DOT) com> wrote

Quote:

I've had this intermittent problem. Couple of times in perhaps last two
weeks.

I'm running d3 on Red Hat Linux. Not sure of the version numbers, but if
this makes a difference, I can post them.

Switch computer on. Goes through memory & SCSI card check OK.

At some point (and I am not quite certain where as I have not been in from
of the screen) stops with the message:

Loading Linux
Uncompressing Linux
CRC error

{then I think it says - I did not write the last line down)

System stopped ....

Pushing the reboot button has got it starting normally.

My questions are:

Is it serious - likely to get worse?

What can a total ignoramus do to sort it - or do I need an expert?
Bearing on mind that my knowledge of Linux is 0.5 on scale of 1 to 10.
Maybe not even that much!

TIA for any help.

Alan Pritchard

#3

Hi Alan,

Quote:

I've had this intermittent problem. Couple of times in perhaps last two
weeks.

I'm running d3 on Red Hat Linux. Not sure of the version numbers, but if
this makes a difference, I can post them.

Switch computer on. Goes through memory & SCSI card check OK.

At some point (and I am not quite certain where as I have not been in from
of the screen) stops with the message:

Loading Linux
Uncompressing Linux
CRC error

{then I think it says - I did not write the last line down)

System stopped ....

Pushing the reboot button has got it starting normally.

The CRC (Cyclic Redundancy Check) error says that while uncompressing
the kernel image the checksums didn't match what was expected. These
are used as a check of the integrity of a compressed archive.

How old is the HDD? Has anything changed in the SCSI card config in the
last couple of weeks? Does a warm boot *always* fix it? I have seen
failing drives which work after having had a bit of a "warm up"
(spinning while things were breaking, then still spinning through the
warm boot).

Quote:

My questions are:

Is it serious

I would consider it as serious as would be the situation in which a
warm boot _doesn't_ fix it - e.g. is this a critical system or a
testbed?

Quote:

- likely to get worse?

If this is the first (obvious) sign of an HDD failure it is likely to
get worse. Are you seeing bad read errors or retrys in /var/log/syslog
(or wherever RH v.??? would log these errors)? If so you should
probably get a good backup and think about replacing the drive(s).

Is your D3 filesystem on the same physical drive? Have you been seeing
GFE's in the D3 filesystem?

Quote:

What can a total ignoramus do to sort it - or do I need an expert?
Bearing on mind that my knowledge of Linux is 0.5 on scale of 1 to 10.
Maybe not even that much!

Log files are your friend. Check your logs and dmesg output. Run any
non-destructive media tests available to you, from your Adaptec (or
whatever) BIOS and the system tests like fsck.

HTH,
-Tom

#4

In article <HvCdnZgYC8PDwsrZnZ2dnUVZ_vSdnZ2d (AT) giganews (DOT) com>,
dfdfg (AT) dfkjdfg (DOT) com (Glen B) wrote:

Quote:

*From:* "Glen B" <dfdfg (AT) dfkjdfg (DOT) com
*Date:* Tue, 2 May 2006 09:32:06 -0400

Could be a failing disk. The CRC error is from the boot image
being uncompressed, which is stored in the master boot record.

run "cat /proc/version" and post your verison info.

I would run e2fsck on the disk manually to check for bad blocks,
especially block zero.

Glen

"Alan Pritchard" <alan.pritchard (AT) gmail (DOT) com> wrote in message
news:memo.20060502142014.4160C (AT) aovq45 (DOT) cix.co.uk...
I've had this intermittent problem. Couple of times in perhaps last
two
weeks.

I'm running d3 on Red Hat Linux. Not sure of the version numbers, but
if
this makes a difference, I can post them.

Switch computer on. Goes through memory & SCSI card check OK.

At some point (and I am not quite certain where as I have not been in
from
of the screen) stops with the message:

Loading Linux
Uncompressing Linux
CRC error

{then I think it says - I did not write the last line down)

System stopped ....

Pushing the reboot button has got it starting normally.

My questions are:

Is it serious - likely to get worse?

What can a total ignoramus do to sort it - or do I need an expert?
Bearing on mind that my knowledge of Linux is 0.5 on scale of 1 to 10.
Maybe not even that much!

TIA for any help.

Alan Pritchard

According to d3's WHICH

d3 IS 7.2.1 linux

and

Unix Information. . . . . Linux;pick0:LINUX;2.4.2-2;#1 Sun Apr 8 20:41:30
EDT 2001;0978156F

I would think that 2001 is probably when the system was set up.

Running cat etc from within d3 gives

!cat proc/version
cat: proc/version: No such file or directory

so I'm probably doing it wrong.

Is 'e2fsck' a Linux program that I can get at from root (which is where I
go after the login: prompt.

Alan Pritchard

Please reply to: alan.pritchard (AT) gmail (DOT) com

#5

In article <1146577314.590441.269360 (AT) i39g2000cwa (DOT) googlegroups.com>,
ted (AT) blackflute (DOT) com (Tom deL) wrote:

Quote:

*From:* "Tom deL" <ted (AT) blackflute (DOT) com
*Date:* 2 May 2006 06:41:54 -0700

Hi Alan,

I've had this intermittent problem. Couple of times in perhaps last
two
weeks.

I'm running d3 on Red Hat Linux. Not sure of the version numbers, but
if
this makes a difference, I can post them.

Switch computer on. Goes through memory & SCSI card check OK.

At some point (and I am not quite certain where as I have not been in
from
of the screen) stops with the message:

Loading Linux
Uncompressing Linux
CRC error

{then I think it says - I did not write the last line down)

System stopped ....

Pushing the reboot button has got it starting normally.

The CRC (Cyclic Redundancy Check) error says that while uncompressing
the kernel image the checksums didn't match what was expected. These
are used as a check of the integrity of a compressed archive.

How old is the HDD? Has anything changed in the SCSI card config in the
last couple of weeks? Does a warm boot *always* fix it? I have seen
failing drives which work after having had a bit of a "warm up"
(spinning while things were breaking, then still spinning through the
warm boot).

My questions are:

Is it serious

I would consider it as serious as would be the situation in which a
warm boot _doesn't_ fix it - e.g. is this a critical system or a
testbed?

- likely to get worse?

If this is the first (obvious) sign of an HDD failure it is likely to
get worse. Are you seeing bad read errors or retrys in /var/log/syslog
(or wherever RH v.??? would log these errors)? If so you should
probably get a good backup and think about replacing the drive(s).

Is your D3 filesystem on the same physical drive? Have you been seeing
GFE's in the D3 filesystem?

What can a total ignoramus do to sort it - or do I need an expert?
Bearing on mind that my knowledge of Linux is 0.5 on scale of 1 to 10.
Maybe not even that much!

Log files are your friend. Check your logs and dmesg output. Run any
non-destructive media tests available to you, from your Adaptec (or
whatever) BIOS and the system tests like fsck.

HTH,
-Tom

The drives (2 of them) are about 5 years old. No to SCSI changes. System
hasn't altered all this time.

Warm boot (so far only 2 occurrences - fairly widely spaced) has always
fixed it.

Mission (and livelihood) critical

How do I get at these logs?

AFAIR all the d3 is on one drive & Linux + a disk-to-disk backup on the
other. No GFEs at all on d3

Does this make it look as through it might be the Linux drive that is
failing, rather than the d3 one.

This could not have come at a worse time. I have major jobs to get out.

Thanks to both of you for the suggestions.

Alan Pritchard

Please reply to: alan.pritchard (AT) gmail (DOT) com

#6

Not too helpful, but the rule of thumb is that a system will go wrong just
when you can't spare any downtime to investigate! In my case, it's always
when I'm on holiday!!!

Simon

--
================================
Simon Verona
Dealer Management Service Ltd
Stewart House
Centurion Business Park
Julian Way
Sheffield
S9 1GD

Tel: 0870 080 2300
Fax: 0870 735 0011

"Alan Pritchard" <alan.pritchard (AT) gmail (DOT) com> wrote

Quote:

In article <1146577314.590441.269360 (AT) i39g2000cwa (DOT) googlegroups.com>,
ted (AT) blackflute (DOT) com (Tom deL) wrote:

*From:* "Tom deL" <ted (AT) blackflute (DOT) com
*Date:* 2 May 2006 06:41:54 -0700

Hi Alan,

I've had this intermittent problem. Couple of times in perhaps last
two
weeks.

I'm running d3 on Red Hat Linux. Not sure of the version numbers, but
if
this makes a difference, I can post them.

Switch computer on. Goes through memory & SCSI card check OK.

At some point (and I am not quite certain where as I have not been in
from
of the screen) stops with the message:

Loading Linux
Uncompressing Linux
CRC error

{then I think it says - I did not write the last line down)

System stopped ....

Pushing the reboot button has got it starting normally.

The CRC (Cyclic Redundancy Check) error says that while uncompressing
the kernel image the checksums didn't match what was expected. These
are used as a check of the integrity of a compressed archive.

How old is the HDD? Has anything changed in the SCSI card config in the
last couple of weeks? Does a warm boot *always* fix it? I have seen
failing drives which work after having had a bit of a "warm up"
(spinning while things were breaking, then still spinning through the
warm boot).

My questions are:

Is it serious

I would consider it as serious as would be the situation in which a
warm boot _doesn't_ fix it - e.g. is this a critical system or a
testbed?

- likely to get worse?

If this is the first (obvious) sign of an HDD failure it is likely to
get worse. Are you seeing bad read errors or retrys in /var/log/syslog
(or wherever RH v.??? would log these errors)? If so you should
probably get a good backup and think about replacing the drive(s).

Is your D3 filesystem on the same physical drive? Have you been seeing
GFE's in the D3 filesystem?

What can a total ignoramus do to sort it - or do I need an expert?
Bearing on mind that my knowledge of Linux is 0.5 on scale of 1 to 10.
Maybe not even that much!

Log files are your friend. Check your logs and dmesg output. Run any
non-destructive media tests available to you, from your Adaptec (or
whatever) BIOS and the system tests like fsck.

HTH,
-Tom

The drives (2 of them) are about 5 years old. No to SCSI changes. System
hasn't altered all this time.

Warm boot (so far only 2 occurrences - fairly widely spaced) has always
fixed it.

Mission (and livelihood) critical

How do I get at these logs?

AFAIR all the d3 is on one drive & Linux + a disk-to-disk backup on the
other. No GFEs at all on d3

Does this make it look as through it might be the Linux drive that is
failing, rather than the d3 one.

This could not have come at a worse time. I have major jobs to get out.

Thanks to both of you for the suggestions.

Alan Pritchard

Please reply to: alan.pritchard (AT) gmail (DOT) com

#7

Hi Alan,

Quote:

Please reply to: alan.pritchard (AT) gmail (DOT) com

Check your e-mail.

#8

In article <4457757f$0$23189$ed2e19e4 (AT) ptn-nntp-reader04 (DOT) plus.net>,
nomail (AT) nomail (DOT) zzz (Simon Verona) wrote:

Quote:

*From:* "Simon Verona" <nomail (AT) nomail (DOT) zzz
*Date:* Tue, 2 May 2006 16:06:32 +0100

Not too helpful, but the rule of thumb is that a system will go wrong
just when you can't spare any downtime to investigate! In my case,
it's always when I'm on holiday!!!

Simon

Yes, I'm convinced there is a special sensor in ALL equipment to monitor
the urgency & then fail if it goes above a limit!!

Alan Pritchard

Please reply to: alan.pritchard (AT) gmail (DOT) com

#9

Simon Verona wrote:

Quote:

Not too helpful, but the rule of thumb is that a system will go wrong
just when you can't spare any downtime to investigate! In my case,
it's always when I'm on holiday!!!

Exactly.

I had a system crash and we couldn't get it to start at the customer's
site. We brought it back to the office and it started fine.

The customer wanted it back right away, so we let them have it. It ran
flawlessly for about another 2 months, then boom. Total crash with
data loss.

These signs should not be ignored.

--
Kevin Powick

#10

In article <HvCdnZgYC8PDwsrZnZ2dnUVZ_vSdnZ2d (AT) giganews (DOT) com>,
dfdfg (AT) dfkjdfg (DOT) com (Glen B) wrote:

Quote:

*From:* "Glen B" <dfdfg (AT) dfkjdfg (DOT) com
*Date:* Tue, 2 May 2006 09:32:06 -0400

Could be a failing disk. The CRC error is from the boot image
being uncompressed, which is stored in the master boot record.

run "cat /proc/version" and post your verison info.

I would run e2fsck on the disk manually to check for bad blocks,
especially block zero.

Glen

"Alan Pritchard" <alan.pritchard (AT) gmail (DOT) com> wrote in message
news:memo.20060502142014.4160C (AT) aovq45 (DOT) cix.co.uk...
I've had this intermittent problem. Couple of times in perhaps last
two
weeks.

I'm running d3 on Red Hat Linux. Not sure of the version numbers, but
if
this makes a difference, I can post them.

Switch computer on. Goes through memory & SCSI card check OK.

At some point (and I am not quite certain where as I have not been in
from
of the screen) stops with the message:

Loading Linux
Uncompressing Linux
CRC error

{then I think it says - I did not write the last line down)

System stopped ....

Pushing the reboot button has got it starting normally.

My questions are:

Is it serious - likely to get worse?

What can a total ignoramus do to sort it - or do I need an expert?
Bearing on mind that my knowledge of Linux is 0.5 on scale of 1 to 10.
Maybe not even that much!

TIA for any help.

Alan Pritchard

Version info from /proc/version =

Linux version 2.4.2-2
Red Hat 7.1 2.96-79

Alan Pritchard

Please reply to: alan.pritchard (AT) gmail (DOT) com