LinuxQuestions.org
Download your favorite Linux distribution at LQ ISO.
Go Back   LinuxQuestions.org > Forums > Linux Forums > Linux - Newbie
User Name
Password
Linux - Newbie This Linux forum is for members that are new to Linux.
Just starting out and have a question? If it is not in the man pages or the how-to's this is the place!

Notices


Reply
  Search this Thread
Old 01-10-2007, 10:08 PM   #1
soniquser
LQ Newbie
 
Registered: Jan 2007
Posts: 16

Rep: Reputation: 0
Help Me!!


Hi all gurus,

Its great to be joining a forum that is really helpful, much informative and a whole lot of kewl people.

I have just started working with a new company here and the systems here are all Linux based. Previously, I was mainly looking after Solaris machines. Its a new challenge for me as the systems here are quite ancient and the criticality for all the systems are really CRITICAL...

Anyway, I have encountered some of this error that has been going on for days. Rebooting the systems just fix the problem temporarily... then it happens again:
EXT3-fs error (device ide0(3,2)) in ext3_reserve_inode_write: IO failure
end_request: I/O error, dev 03:02 (hda), sector 0
end_request: I/O error, dev 03:02 (hda), sector 1835040
EXT3-fs error (device ide0(3,2)): ext3_get_inode_loc: unable to read inode block - inode=102377, block=229380
end_request: I/O error, dev 03:02 (hda), sector 0
EXT3-fs error (device ide0(3,2)) in ext3_reserve_inode_write: IO failure
end_request: I/O error, dev 03:02 (hda), sector 0
end_request: I/O error, dev 03:02 (hda), sector 1835040
EXT3-fs error (device ide0(3,2)): ext3_get_inode_loc: unable to read inode block - inode=102376, block=229380
end_request: I/O error, dev 03:02 (hda), sector 0
EXT3-fs error (device ide0(3,2)) in ext3_reserve_inode_write: IO failure
end_request: I/O error, dev 03:02 (hda), sector 0
end_request: I/O error, dev 03:05 (hda), sector 3145776
EXT3-fs error (device ide0(3,5)): ext3_get_inode_loc: unable to read inode block - inode=192477, block=393222
end_request: I/O error, dev 03:05 (hda), sector 0
EXT3-fs error (device ide0(3,5)) in ext3_reserve_inode_write: IO failure
end_request: I/O error, dev 03:05 (hda), sector 0

I believe the harddisk is at its end of life...
What is the best approach to resolve the issue?

Please assist...

Warm Regards,
Soniquser
 
Old 01-10-2007, 11:32 PM   #2
PatrickNew
Senior Member
 
Registered: Jan 2006
Location: Charleston, SC, USA
Distribution: Debian, Gentoo, Ubuntu, RHEL
Posts: 1,148
Blog Entries: 1

Rep: Reputation: 48
It looks like a dying harddrive. I guess the next step is to replace the harddisk.
 
Old 01-11-2007, 01:30 AM   #3
Wim Sturkenboom
Senior Member
 
Registered: Jan 2005
Location: Roodepoort, South Africa
Distribution: Slackware 10.1/10.2/12, Ubuntu 12.04, Crunchbang Statler
Posts: 3,786

Rep: Reputation: 282Reputation: 282Reputation: 282
No answer to the question, but please use a better title that covers the topic.
Click the edit button to edit your post, click advanced and you can change it.
 
Old 01-11-2007, 05:04 AM   #4
pixellany
LQ Veteran
 
Registered: Nov 2005
Location: Annapolis, MD
Distribution: Arch/XFCE
Posts: 17,802

Rep: Reputation: 738Reputation: 738Reputation: 738Reputation: 738Reputation: 738Reputation: 738Reputation: 738
Quote:
Its great to be joining a forum that is really helpful, much informative and a whole lot of kewl people.

I have just started working with a new company here and the systems here are all Linux based. Previously, I was mainly looking after Solaris machines. Its a new challenge for me as the systems here are quite ancient and the criticality for all the systems are really CRITICAL...
"kewl"---hmmm. My spellchecker offers "mewl" (maybe like meow??) and "keel". No help there.....
Google gives a whole bunch of options, but it appears that "kewl" in some dialects actually is supposed to mean "cool"......

All seriousness aside, this looks like a problem:
"the systems here are quite ancient and the criticality for all the systems are really CRITICAL" Are you being asked to band-aid a bunch of tired, failing hardware? I hope you don't wind up being blamed for someone's data being lost. Your best long-term strategy may be to counsel the company on what they need to do to their systems. If they don't want to do it, it may be time to look around.....
 
Old 01-11-2007, 05:07 AM   #5
billymayday
LQ Guru
 
Registered: Mar 2006
Location: Sydney, Australia
Distribution: Fedora, CentOS, OpenSuse, Slack, Gentoo, Debian, Arch, PCBSD
Posts: 6,678

Rep: Reputation: 122Reputation: 122
C'mon, lay off the guy - he's got a problem. Sure the title sucks, but at least he needs some help it isn't just "how do I change fonts" type question
 
Old 01-11-2007, 06:10 AM   #6
IndyGunFreak
Senior Member
 
Registered: Aug 2003
Location: Indpls
Distribution: Laptops: Linux Mint 18 XFCE, Debian Jessie XFCE, NAS: OpenMediaVault
Posts: 1,303

Rep: Reputation: 56
Quote:
Originally Posted by billymayday
C'mon, lay off the guy - he's got a problem. Sure the title sucks, but at least he needs some help it isn't just "how do I change fonts" type question
It says right by the subject, while you type a new post, not to use subjects like Help Me, so when you do it, you get what you deserve.

IGF
 
Old 01-11-2007, 06:24 AM   #7
b0uncer
LQ Guru
 
Registered: Aug 2003
Distribution: CentOS, OS X
Posts: 5,131

Rep: Reputation: Disabled
Quote:
It says right by the subject, while you type a new post, not to use subjects like Help Me, so when you do it, you get what you deserve.
Actually I don't think the purpose of LQ is to get or give "what you deserve". If you feel so, maybe it's time to re-read the official announcements and docs about how you were supposed to act in here. Even if somebody had a bad subject, used odd words like 'kewl', couldn't write perfect english or more, it's not a reason to start crying about it and not help with the actual problem. I'm confident you can read the actual problem from the post despite of the "bad topic" (or if you can't, then why bother posting?)

Anyway. Soniquser: if you consider the hardware old, harddisks probably ('probably' is enough if you're wise; don't wait for a disaster to happen) failing and so on, your first thing is to inform the company about these things. And remind them what could happen if nothing was done; ask them how valuable the data is. Second thing is to find the most critical errors, probably harddisks and backups, and start dealing with them; minor 'problems' like old hardware can probably wait until you have made sure backups exist and planned your moves to switch the old harddisks to new ones if it's needed. Don't hesitate to start dealing with the problem, waiting probably won't fix it; your first and greatest thing to worry about is the backup system: is it ok, are backups working (test them if you can, to see if they are working) and so on. Without backups the company might have some bad times coming, with backups your worries are not that bad.
 
Old 01-11-2007, 11:32 AM   #8
PatrickNew
Senior Member
 
Registered: Jan 2006
Location: Charleston, SC, USA
Distribution: Debian, Gentoo, Ubuntu, RHEL
Posts: 1,148
Blog Entries: 1

Rep: Reputation: 48
On the etiquette question:
(1) The point of telling a guy who writes a bad question isn't to screw him over, it's so that he hopefully won't write bad questions in the future, improving the overall readability of the board. (2) On his header I notice "Posts: 1", so he's a newb to the board. He needs to be told, but we don't need to flame him. Lot's of people ignore the fine print on a web page, next to fields because they seldom include important information.

To the thread author: I'm sorry you got flamed. However, many people here spend a lot of time answering questions and would really appreciate more descriptive titles and clear text. If you make them happy, they answer faster, so it's just a good idea. No harm done, just don't do it again.

On a technical question: If a hard drive is giving buggy output, how does one verify it? What does one have to compare their backup against? Do you just make multiple backups and look at their differences?
 
Old 01-11-2007, 11:56 PM   #9
Wim Sturkenboom
Senior Member
 
Registered: Jan 2005
Location: Roodepoort, South Africa
Distribution: Slackware 10.1/10.2/12, Ubuntu 12.04, Crunchbang Statler
Posts: 3,786

Rep: Reputation: 282Reputation: 282Reputation: 282
I did not intend to be nasty. If I was, I would not have advised how to change it.

With regards to the backup advised above: make sure that your backup is OK. Backing up already corrupted data and/or mistakes in the backup make it quite useless. So extract the backup on another machine to check.
It happend to one of my previous employers once and the company lost over 50% of the emails of all employees due to corrupted backups.
 
Old 01-12-2007, 12:31 AM   #10
JimBass
Senior Member
 
Registered: Oct 2003
Location: New York City
Distribution: Debian Sid 2.6.32
Posts: 2,100

Rep: Reputation: 49
I think the combination of "safest & easiest" would be to get a large to huge USB disk, and copy the data onto that. Where you'll have a big problem is if the drives on this machine are large, or it is part of a farm of servers, where all of the info is frequently changing. If one of the things you're trying to copy is a big database, you may be in major trouble.

I would check the /etc/fstab, and see how many partitions are on the machine with the bad disk(s). Hopefully as a server, it has partitions for every root directory, and only one of them is going bad. The command df will give you simple, human readable output on how much space you have used on each of the partitions. Say for example that the /var partition is one of the bad ones. Say /var has 50 Gb of space, but is only using 15 Gb of it. In that case, plug in the external USB drive, format it with some linux FS (ext3 is fine, so are many others, but not fat32), and basically copy the whole /var partition onto the USB.

Personally if I were to do that, I would pull the server out of production, and do it with a live disk, like knoppix or whatever. You could either straight make a copy of the data, or make a tar.gz of it on the USB disk, whatever works. Before you do anything else, move the USB to another machine, mount it, and see if the data came off cleanly. If you do make a tar, you'll want to do something like --same-owner to keep permissions where they should be.

Unfortunately, You'll probably find that being a server, one of the scsi disks is going bad, and that's not an easy thing to replace. You'll need to figure out every partition that is on that particular disk, and get all of them off. A simple solution would be to copy all the data in trouble to the USB, put a new SCSI in place of the bad, then again use knoppix to copy off the data, then reboot. It can get really hairy though if it is SCSI under RAID, depending on the type of RAID. RAID 1 has a mirror copy, no problem, let the controller rebuild. RAID 5 and 10 you can let the controller rebuild. If it is RAID 0, you're SOL. Copy and pray!

Obviously I'm thinking this is a big proper server. It could just be a box called a server, running on one physical ide drive! We'll need a bit more info before we can give you an absolute, "do this, then that, then the other." So please, do let us see the fstab, let us know physically what type of disks were talking about, how many there are, and how many are in bad shape. Also, can this box be pulled down and not bring the entire rest of the operation to a screeching halt?

Peace,
JimBass

Last edited by JimBass; 01-12-2007 at 12:36 AM.
 
Old 01-16-2007, 09:40 PM   #11
soniquser
LQ Newbie
 
Registered: Jan 2007
Posts: 16

Original Poster
Rep: Reputation: 0
Hi guys,

Firstly, I would like to apologize for the terrible entrance... and thanks to all for the valueable advices.

I will make sure that the next time that I want to post anything the subject will be more appropriate.

Well, the harddisk was replaced and thankfully almost all data was recovered.

Thanks all, again.
 
Old 01-16-2007, 09:43 PM   #12
PatrickNew
Senior Member
 
Registered: Jan 2006
Location: Charleston, SC, USA
Distribution: Debian, Gentoo, Ubuntu, RHEL
Posts: 1,148
Blog Entries: 1

Rep: Reputation: 48
Glad to hear it.
 
Old 01-16-2007, 09:46 PM   #13
soniquser
LQ Newbie
 
Registered: Jan 2007
Posts: 16

Original Poster
Rep: Reputation: 0
Hi guys,

Firstly, I would like to apologize for the terrible entrance... and thanks to all for the valueable advices.

I will make sure that the next time that I want to post anything the subject will be more appropriate.

Well, the harddisk was replaced and thankfully almost all data was recovered.

Thanks all, again.
 
  


Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off




All times are GMT -5. The time now is 02:51 AM.

Main Menu
Advertisement
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
Facebook: linuxquestions Google+: linuxquestions
Open Source Consulting | Domain Registration