[HPADM] [Summary] drive failure

From: Hutchins, Greg (ghutchins2@selfregional.org)
Date: Wed Aug 18 2004 - 10:27:45 EDT


Once again,

Thanks to all those who replied! The help on this list has been great!
The general consensus was that the box did not have to go down. Thanks
to Mark who pointed me to a document on itrc that gave step-by-step
instructions on replacing the drive. However, the HP rep did not feel
comfortable swapping it out while the box was up, so we rebooted into
maintenance mode and did the following:

 

vgcfgrestore -n /dev/vg00 /dev/rdsk/cxtxdx

vgchange -a y /dev/vg00

vgsync vg00

mkboot /dev/rdsk/cxtxdx

mkboot -a "hpux"

 

 

Afterwards, we were able to boot off of the replaced drive and all the
logical volumes are synced again (instead of stale)


attached mail follows:


You need to un-mirror the dead disk (lvreduce -m 0 /dev/vg00/lvol
/dev/dsk/c...)
This needs to be done on all lvols that are on the root disk. I don't know
if the
disk are hotswap but you do need to make sure the LVM system has all the
extents marked stale, or un-mirror the bad disk. Replace the disk, then use
lvextend -m 1 to re-add the disk to the VG. When using lvextend -m 1, the
command will also sync the new disk lvols.
 

--
Bill
-----Original Message-----
From: hpux-admin-owner@DutchWorks.nl [mailto:hpux-admin-owner@DutchWorks.nl]
On Behalf Of Hutchins, Greg
Sent: Tuesday, August 17, 2004 12:50 PM
To: hpux-admin@DutchWorks.nl
Subject: [HPADM] drive failure
Hi all,
We have an rp7410 with 2 internal 36.4GB drives.  These drives host vg00
which is mirrored.  One of the drives has failed and we have a replacement
coming.  I was told that the system has to go down for about an hour to
change out this drive since it is in the root volume group.  Is this not
something that is hot swappable and what is involved with swapping a drive
out?  I know with our AIX boxes, we have been able to swap these out without
bringing the system down.  Thanks in advance for any replies.

attached mail follows:


you do not need to shutdown the system ...you can do it while the OS is
online regardless of what vol. grp. it is
 
just make sure before you pull out the bad drive that LVM has marked the PEs
on this drive as stale and that the PV is marked as bad/unavailable
 
HP can give u the steps to resync to the new drive you will put in its place
 
marc

-----Original Message-----
From: Hutchins, Greg [mailto:ghutchins2@selfregional.org]
Sent: Tuesday, August 17, 2004 9:50 AM
To: hpux-admin@DutchWorks.nl
Subject: [HPADM] drive failure

Hi all,

We have an rp7410 with 2 internal 36.4GB drives. These drives host vg00
which is mirrored. One of the drives has failed and we have a replacement
coming. I was told that the system has to go down for about an hour to
change out this drive since it is in the root volume group. Is this not
something that is hot swappable and what is involved with swapping a drive
out? I know with our AIX boxes, we have been able to swap these out without
bringing the system down. Thanks in advance for any replies.


attached mail follows:


Greg,

I'm far from an expert, but I have used the following procedure before.

1. Replace the drive physically.
2. vgcfgrestore -n /dev/vg00 /dev/rdsk/cXtXdX (replace the 'X's)
3. vgchange -a y /dev/vg00
4. vgsync vg00
5. mkboot /dev/rdsk/cXtXdX (makes it bootable)
6. mkboot -a "hpux (;0)/stand/vmunix" /dev/rdsk/cXtXdX
    (tell the LIF what file to boot from by default)
7. lvlnboot -Rv (tells vg to be root/boot vg)

To verify your LIF, do lifcp /dev/dsk/cXtXdX:AUTO -

HTH,

Ed

> Hi all,
>
> We have an rp7410 with 2 internal 36.4GB drives. These drives host vg00
> which is mirrored. One of the drives has failed and we have a
> replacement coming. I was told that the system has to go down for about
> an hour to change out this drive since it is in the root volume group.
> Is this not something that is hot swappable and what is involved with
> swapping a drive out? I know with our AIX boxes, we have been able to
> swap these out without bringing the system down. Thanks in advance for
> any replies.
>
>

0xed


attached mail follows:


No. The system doesn't have to go down.

1. Replace the drive.
2. Do a pvcreate -B on the drive (rdsk).
3. vgcfgrestore -n /dev/vg00 drivepath (rdsk)
4. vgchange -a y vg00 (should start the sync process, otherwise use vgsync)

-Roger

"Hutchins, Greg" <ghutchins2@selfregional.org>@dutchworks.nl on 17 Aug 2004
12:50

Sent by: hpux-admin-owner@dutchworks.nl

To: <hpux-admin@dutchworks.nl>
Subject: [HPADM] drive failure

****************************************************************************

 This message originated from the Internet. Its originator may or
 may not be who they claim to be and the information contained in
 the message and any attachments may or may not be accurate.
****************************************************************************

Hi all,

We have an rp7410 with 2 internal 36.4GB drives. These drives host vg00
which is mirrored. One of the drives has failed and we have a
replacement coming. I was told that the system has to go down for about
an hour to change out this drive since it is in the root volume group.
Is this not something that is hot swappable and what is involved with
swapping a drive out? I know with our AIX boxes, we have been able to
swap these out without bringing the system down. Thanks in advance for
any replies.

(See attached file: C.htm)

************************************************************************
 This E-mail is confidential. It may also be legally privileged. If you
 are not the addressee you may not copy, forward, disclose or use any
 part of it. If you have received this message in error, please delete
 it and all copies from your system and notify the sender immediately
 by return E-mail.

 Internet communications cannot be guaranteed to be timely, secure,
 error or virus-free. The sender does not accept liability for any
 errors or omissions.
************************************************************************



attached mail follows:


Greg-

The drives are hot "pluggable", not hot swappable...there's a difference.
Hot plug means you don't have to shut down to physically switch the drives,
but then you do have to configure, which probably requires a reboot (hence,
the shutdown). And software mirroring I believe also requires some
reconfig work. We always hardware mirror so I'm not that familiar with
software mirroring.

If you want to remove the "shutdown", then I think you have to hardware
mirror, which requires hardware mounted below the CPU. Let me know if you
want more info about this.

Best,

John Lee
Vaske Computer Solutions

We sell, service, support and integrate HP 3000/9000, new and old, since 1988

At 12:50 PM 8/17/04 -0400, you wrote:
Hi all,

We have an rp7410 with 2 internal 36.4GB drives. These drives host vg00
which is mirrored. One of the drives has failed and we have a replacement
coming. I was told that the system has to go down for about an hour to
change out this drive since it is in the root volume group. Is this not
something that is hot swappable and what is involved with swapping a drive
out? I know with our AIX boxes, we have been able to swap these out
without bringing the system down. Thanks in advance for any replies.


attached mail follows:


X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
Received: from emailfilter.selfregional.org ([172.16.2.97]) by exchange.selfregional.org with Microsoft SMTPSVC(5.0.2195.6713); Tue, 17 Aug 2004 15:02:27 -0400
MIME-Version: 1.0
Content-Type: multipart/alternative;
        boundary="----_=_NextPart_009_01C4848C.BCBCE380"
Received: from omr2.netsolmail.com [216.168.230.163] by emailfilter.selfregional.org - SurfControl E-mail Filter (4.7); Tue, 17 Aug 2004 15:02:21 -0400
Received: from ms7.netsolmail.com (IDENT:mirapoint@[216.168.230.174]) by omr2.netsolmail.com (8.12.10/8.12.10) with ESMTP id i7HJ2G6i026062 for <ghutchins2@selfregional.org>; Tue, 17 Aug 2004 15:02:17 -0400 (EDT)
Received: from ms7.netsolmail.com (localhost.netsolmail.com [127.0.0.1]) by ms7.netsolmail.com (Mirapoint Messaging Server MOS 3.2.2-GA) with ESMTP id CEU18018; Tue, 17 Aug 2004 15:02:12 -0400 (EDT)
Received: from 162.136.193.1 by ms7.netsolmail.com (Mirapoint Messaging Server MOS 3.2.2-GA) with HTTP/1.1; Tue, 17 Aug 2004 15:02:12 -0400
Content-Class: urn:content-classes:message
Subject: Re: [HPADM] drive failure
Date: Tue, 17 Aug 2004 15:02:12 -0400
Message-ID: <200408171902.CEU18018@ms7.netsolmail.com>
X-MS-Has-Attach:
X-MS-TNEF-Correlator:
Thread-Topic: [HPADM] drive failure
thread-index: AcSEjL1RuW9I5RyaT3qyotNLvni3tQ==
From: "Mark Graham" <mark_graham@unxguys.net>
To: "Hutchins, Greg" <ghutchins2@selfregional.org>
Cc: "Hutchins, Greg" <ghutchins2@selfregional.org>
Reply-To: <mark_graham@unxguys.net>

This is a multi-part message in MIME format.

------_=_NextPart_009_01C4848C.BCBCE380
Content-Type: text/plain;
        charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Greg,

Try ITRC doc LVMKBRC00006111 - an excellent document for=20
replacing a hot swap, mirrored disk...

Mark

---- Original message ----
>Date: Tue, 17 Aug 2004 12:50:06 -0400
>From: "Hutchins, Greg" <ghutchins2@selfregional.org> =20
>Subject: [HPADM] drive failure =20
>To: <hpux-admin@DutchWorks.nl>
>
> Hi all,
>
> We have an rp7410 with 2 internal 36.4GB drives.=20
> These drives host vg00 which is mirrored. One of
> the drives has failed and we have a replacement
> coming. I was told that the system has to go down
> for about an hour to change out this drive since it
> is in the root volume group. Is this not something
> that is hot swappable and what is involved with
> swapping a drive out? I know with our AIX boxes, we
> have been able to swap these out without bringing
> the system down. Thanks in advance for any replies.
Mark Graham
Positive Outcomes Desired

Graham/Milora, Inc. dba GMI Consulting (UNXGUYS)
www.unxguys.com
504-833-4916 (Office)
860-843-3520 (Client Office)
860-379-1571 (Corporate Apartment)
504-982-3278 (PCS)
504-616-5905 (Car)
504-833-5937 (Fax)
mark_graham@unxguys.com (alternate e-mail)

------_=_NextPart_009_01C4848C.BCBCE380
Content-Type: text/html;
        charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
6.0.6249.1">
<TITLE>Re: [HPADM] drive failure</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->

<P><FONT SIZE=3D2>Greg,<BR>
<BR>
Try ITRC doc LVMKBRC00006111 - an excellent document for<BR>
replacing a hot swap, mirrored disk...<BR>
<BR>
Mark<BR>
<BR>
---- Original message ----<BR>
&gt;Date: Tue, 17 Aug 2004 12:50:06 -0400<BR>
&gt;From: &quot;Hutchins, Greg&quot; =
&lt;ghutchins2@selfregional.org&gt;&nbsp;<BR>
&gt;Subject: [HPADM] drive failure&nbsp;<BR>
&gt;To: &lt;hpux-admin@DutchWorks.nl&gt;<BR>
&gt;<BR>
&gt;&nbsp;&nbsp; Hi all,<BR>
&gt;<BR>
&gt;&nbsp;&nbsp; We have an rp7410 with 2 internal 36.4GB drives.<BR>
&gt;&nbsp;&nbsp; These drives host vg00 which is mirrored.&nbsp; One =
of<BR>
&gt;&nbsp;&nbsp; the drives has failed and we have a replacement<BR>
&gt;&nbsp;&nbsp; coming.&nbsp; I was told that the system has to go =
down<BR>
&gt;&nbsp;&nbsp; for about an hour to change out this drive since it<BR>
&gt;&nbsp;&nbsp; is in the root volume group.&nbsp; Is this not =
something<BR>
&gt;&nbsp;&nbsp; that is hot swappable and what is involved with<BR>
&gt;&nbsp;&nbsp; swapping a drive out?&nbsp; I know with our AIX boxes, =
we<BR>
&gt;&nbsp;&nbsp; have been able to swap these out without bringing<BR>
&gt;&nbsp;&nbsp; the system down.&nbsp; Thanks in advance for any =
replies.<BR>
Mark Graham<BR>
Positive Outcomes Desired<BR>
<BR>
Graham/Milora, Inc. dba GMI Consulting (UNXGUYS)<BR>
www.unxguys.com<BR>
504-833-4916 (Office)<BR>
860-843-3520 (Client Office)<BR>
860-379-1571 (Corporate Apartment)<BR>
504-982-3278 (PCS)<BR>
504-616-5905 (Car)<BR>
504-833-5937 (Fax)<BR>
mark_graham@unxguys.com (alternate e-mail)<BR>
<BR>
<BR>
</FONT>
</P>

</BODY>
</HTML>
------_=_NextPart_009_01C4848C.BCBCE380--

--
             ---> Please post QUESTIONS and SUMMARIES only!! <---
        To subscribe/unsubscribe to this list, contact majordomo@dutchworks.nl
       Name: hpux-admin@dutchworks.nl     Owner: owner-hpux-admin@dutchworks.nl
 
 Archives:  ftp.dutchworks.nl:/pub/digests/hpux-admin       (FTP, browse only)
            http://www.dutchworks.nl/htbin/hpsysadmin   (Web, browse & search)



This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 11:02:42 EDT