[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [Linux-cluster] Linux-cluster Digest, Vol 77, Issue 5



 
Hi

It seems you are using hostname of cluster nodes at the place of
hostname of ilo (ILO should have separate ip and hostname in DNS)

In below config is node1.drctmb.com assigned as hostname of node or the
hostname of ILO device? It should be hostname of ilo device..


<fencedevices>
> <fencedevice agent="fence_ilo" hostname="node1.drctmb.com"
> login="root" name="NODE1" passwd="redhat123"/>
> <fencedevice agent="fence_ilo" hostname="node2.drctmb.com"
> login="root" name="NODE2" passwd="redhat123"/>
> </fencedevices>

Thanks
Anoop

-----Original Message-----
From: linux-cluster-bounces redhat com
[mailto:linux-cluster-bounces redhat com] On Behalf Of
linux-cluster-request redhat com
Sent: Thursday, September 09, 2010 12:00 PM
To: linux-cluster redhat com
Subject: Linux-cluster Digest, Vol 77, Issue 5

Send Linux-cluster mailing list submissions to
	linux-cluster redhat com

To subscribe or unsubscribe via the World Wide Web, visit
	https://www.redhat.com/mailman/listinfo/linux-cluster
or, via email, send a message with subject or body 'help' to
	linux-cluster-request redhat com

You can reach the person managing the list at
	linux-cluster-owner redhat com

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Linux-cluster digest..."


Today's Topics:

   1. Re: need help - Fencing problem (ESGLinux)
   2. Re: need help - Fencing problem (rhurst bidmc harvard edu)
   3. Re: need help - Fencing problem (Nehemias Jahcob)
   4. Re: need help - Fencing problem (Ben Turner)


----------------------------------------------------------------------

Message: 1
Date: Thu, 9 Sep 2010 10:51:47 +0200
From: ESGLinux <esggrupos gmail com>
To: linux clustering <linux-cluster redhat com>
Subject: Re: [Linux-cluster] need help - Fencing problem
Message-ID:
	<AANLkTinvwGjTYc01AgdghwTszSm2j5TJXyw+nEmEwGFe mail gmail com>
Content-Type: text/plain; charset="iso-8859-1"

Hi,

the only reason was that when I used as shared the speed of this device
was
very very low. Marked it as non-shared it works fine. I don?t know the
reason. It was a try-error test,

Greetings,

ESG

2010/9/9 Jankowski, Chris <Chris Jankowski hp com>

>  Why did you have to set iLO as non-shared?
>
> Thank and regards,
>
> Chris
>
>  ------------------------------
> *From:* linux-cluster-bounces redhat com [mailto:
> linux-cluster-bounces redhat com] *On Behalf Of *ESGLinux
> *Sent:* Wednesday, 8 September 2010 22:57
> *To:* linux clustering
>
> *Subject:* Re: [Linux-cluster] need help - Fencing problem
>
> Hello,
>
> Have you configured the iLO devices entering in the BIOS?
>
> I remenber I have to set up the user/pass in the iLO and marked the
iLo as
> not shared
>
>
> HTH,
>
> ESG
>
> 2010/9/8 Girish Prajapati <girishpati yahoo com>
>
>>  Hello Everybody,
>> i am having problem of fencing a cluster node  let me explain
indetail :
>> I have installed RHEL 5.4 on  HP Prolaint DL280 G5 servers and iLO
2as
>> fencing device. Am managing cluster through Luci - (Conga). itseems
>> everything is working fine. I can reboot cluster nodes through Luci
and
>> service get transfer to another node. After rebooting node connect to
>> cluster automatically without any error.
>> Problem is i can not do Fence this node through Luci, when i try to
fence
>> any node i get following error :
>>
>> Sep  8 14:51:16 node2 fence_node[9106]: agent "fence_ilo" reports:
Unable
>> to connect/login to fencing device
>> Sep  8 14:51:16 node2 fence_node[9106]: Fence of "node1.drctmb.com"
was
>> unsuccessful
>>
>> my iLO license is : iLO 2 Advanced Evaluation
>> Do i need to have  license of iLO or there is problem in
configuration of
>> cluster ?
>> how i can check cluster log in details.
>>
>> Appreciate your help.
>> Thank you in advance.
>>
>> Regards,
>> Girishkumar R Prajapati
>>
>>
>> --
>> Linux-cluster mailing list
>> Linux-cluster redhat com
>> https://www.redhat.com/mailman/listinfo/linux-cluster
>>
>
>
> --
> Linux-cluster mailing list
> Linux-cluster redhat com
> https://www.redhat.com/mailman/listinfo/linux-cluster
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<https://www.redhat.com/archives/linux-cluster/attachments/20100909/4398
a868/attachment.html>

------------------------------

Message: 2
Date: Thu, 9 Sep 2010 09:34:20 -0400
From: <rhurst bidmc harvard edu>
To: <linux-cluster redhat com>
Subject: Re: [Linux-cluster] need help - Fencing problem
Message-ID:
	
<50168EC934B8D64AA8D8DD37F840F3DE05640628E6 EVS2CCR its caregroup org>
Content-Type: text/plain; charset="us-ascii"

For what it is worth, our experiences with HP iLO management cards:

iLO found on G1 servers does not need to be licensed, AFAIK, it does not
have the option to do so anyways.

iLO2 found on G2 and beyond does not need to be licensed either, if you
are only using it as a fencing device.  We licensed all of ours, because
it enabled useful KVM with remote media capabilities that are superior
than our Raritan KVM infrastructure.

Both management cards should have their firmware updated -- they were
both problematic to us as factory-shipped, but applying their update
packs allowed them to work as advertised.

Also, can't you add "-v" for verbose output and also something like "-D
/tmp/fence.out" to save debugging info to an output file?  It might help
some to see where exactly the failure is occuring.  Good luck.

________________________________
From: linux-cluster-bounces redhat com
[mailto:linux-cluster-bounces redhat com] On Behalf Of Girish Prajapati
Sent: Wednesday, September 08, 2010 6:06 AM
To: Linux-cluster redhat com
Subject: [Linux-cluster] need help - Fencing problem

Hello Everybody,
i am having problem of fencing a cluster node  let me explain indetail :
I have installed RHEL 5.4 on  HP Prolaint DL280 G5 servers and iLO 2as
fencing device. Am managing cluster through Luci - (Conga). itseems
everything is working fine. I can reboot cluster nodes through Luci and
service get transfer to another node. After rebooting node connect to
cluster automatically without any error.
Problem is i can not do Fence this node through Luci, when i try to
fence any node i get following error :

Sep  8 14:51:16 node2 fence_node[9106]: agent "fence_ilo" reports:
Unable to connect/login to fencing device
Sep  8 14:51:16 node2 fence_node[9106]: Fence of "node1.drctmb.com" was
unsuccessful

my iLO license is : iLO 2 Advanced Evaluation
Do i need to have  license of iLO or there is problem in configuration
of cluster ?
how i can check cluster log in details.

Appreciate your help.
Thank you in advance.

Regards,
Girishkumar R Prajapati

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<https://www.redhat.com/archives/linux-cluster/attachments/20100909/d58c
6bb4/attachment.html>

------------------------------

Message: 3
Date: Thu, 9 Sep 2010 10:18:31 -0400
From: Nehemias Jahcob <nehemiasjahcob gmail com>
To: linux clustering <linux-cluster redhat com>
Subject: Re: [Linux-cluster] need help - Fencing problem
Message-ID:
	<AANLkTim-nS3c8e67kPycd-u0XFOMERR8EJorG6+xHn4M mail gmail com>
Content-Type: text/plain; charset="iso-8859-1"

1. ) You can increase the verbosity level for troubleshooting??
   <cluster alias="girish" config_version="*n+1*" name="girish">
----
  <fence_daemon clean_start="0" post_fail_delay="0" post_join_delay="3"
*
log_level="7"/>*
  <rm *log_level="7"*>
-----
#ccs_tool update /etc/cluster/cluster.conf

Copy-paste /var/log/messages


2.) What version of PSP you have installed??

3.) If  nothing works, I recommend using fence_ipmi

Greetings!




2010/9/9 <rhurst bidmc harvard edu>

>  For what it is worth, our experiences with HP iLO management cards:
>
> iLO found on G1 servers does not need to be licensed, AFAIK, it does
not
> have the option to do so anyways.
>
> iLO2 found on G2 and beyond does not need to be licensed either, if
you are
> only using it as a fencing device.  We licensed all of ours, because
it
> enabled useful KVM with remote media capabilities that are superior
than our
> Raritan KVM infrastructure.
>
> Both management cards should have their firmware updated -- they were
both
> problematic to us as factory-shipped, but applying their update
> packs allowed them to work as advertised.
>
> Also, can't you add "-v" for verbose output and also something like
"-D
> /tmp/fence.out" to save debugging info to an output file?  It might
help
> some to see where exactly the failure is occuring.  Good luck.
>
>  ------------------------------
> *From:* linux-cluster-bounces redhat com [mailto:
> linux-cluster-bounces redhat com] *On Behalf Of *Girish Prajapati
> *Sent:* Wednesday, September 08, 2010 6:06 AM
> *To:* Linux-cluster redhat com
> *Subject:* [Linux-cluster] need help - Fencing problem
>
>  Hello Everybody,
> i am having problem of fencing a cluster node  let me explain indetail
:
> I have installed RHEL 5.4 on  HP Prolaint DL280 G5 servers and iLO 2as
> fencing device. Am managing cluster through Luci - (Conga). itseems
> everything is working fine. I can reboot cluster nodes through Luci
and
> service get transfer to another node. After rebooting node connect to
> cluster automatically without any error.
> Problem is i can not do Fence this node through Luci, when i try to
fence
> any node i get following error :
>
> Sep  8 14:51:16 node2 fence_node[9106]: agent "fence_ilo" reports:
Unable
> to connect/login to fencing device
> Sep  8 14:51:16 node2 fence_node[9106]: Fence of "node1.drctmb.com"
was
> unsuccessful
>
> my iLO license is : iLO 2 Advanced Evaluation
> Do i need to have  license of iLO or there is problem in configuration
of
> cluster ?
> how i can check cluster log in details.
>
> Appreciate your help.
> Thank you in advance.
>
> Regards,
> Girishkumar R Prajapati
>
>
> --
> Linux-cluster mailing list
> Linux-cluster redhat com
> https://www.redhat.com/mailman/listinfo/linux-cluster
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<https://www.redhat.com/archives/linux-cluster/attachments/20100909/ba94
0c01/attachment.html>

------------------------------

Message: 4
Date: Thu, 9 Sep 2010 11:58:45 -0400 (EDT)
From: Ben Turner <bturner redhat com>
To: linux clustering <linux-cluster redhat com>
Subject: Re: [Linux-cluster] need help - Fencing problem
Message-ID:
	
<155361964 174311284047925612 JavaMail root zmail07 collab prod int phx2
.redhat.com>
	
Content-Type: text/plain; charset=utf-8

Judging from:

"Sep 8 14:51:16 node2 fence_node[9106]: agent "fence_ilo" reports:
Unable to connect/login to fencing device"

Chances are you are not using the correct username/password/IP or the
ilo is not configured for telnet logins.  Try the following:

1.  Login to the ilo via telnet from the command line.  Be sure to use
the username/password/IP you have in cluster.conf.

2.  If that is successful try:

# fence_ilo -v -a "Ilo IP from cluster.conf" -l "Ilo user from
cluster.conf" -p "Ilo passwd from cluster.conf" -o status

The -v will display exactly what the fence agent sees and is very useful
for debugging failing fences.  If the status fails send me the output.

3.  If the fence_ilo successful try:

# fence_node <node name from cluster.conf>

If all 3 are successful then fencing is setup properly and there may be
a problem running it from Luci, if any of the 3 fail post the error back
to the list and I'll look at it.

-Ben





----- "Girish Prajapati" <girishpati yahoo com> wrote:

> Hello,
> i can run following command successfully from another node but still
> getting same error message :
> 
> fence_ilo -a "Ilo IP" -l "Ilo user" -p "Ilo passwd" -o reboot
> 
> Sep 9 14:37:00 node2 openais[2904]: [CLM ] Members Joined:
> Sep 9 14:37:00 node2 openais[2904]: [SYNC ] This node is within the
> primary component and will provide service.
> Sep 9 14:37:00 node2 openais[2904]: [TOTEM] entering OPERATIONAL
> state.
> Sep 9 14:37:00 node2 openais[2904]: [CLM ] got nodejoin message
> 192.168.0.28
> Sep 9 14:37:00 node2 openais[2904]: [CPG ] got joinlist message from
> node 1
> Sep 9 14:37:00 node2 fenced[2923]: node1.drctmb.com not a cluster
> member after 0 sec post_fail_delay
> Sep 9 14:37:00 node2 fenced[2923]: fencing node "node1.drctmb.com"
> Sep 9 14:37:10 node2 fenced[2923]: agent "fence_ilo" reports: Unable
> to connect/login to fencing device
> Sep 9 14:37:10 node2 fenced[2923]: fence "node1.drctmb.com" failed
> Sep 9 14:37:15 node2 fenced[2923]: fencing node "node1.drctmb.com"
> Sep 9 14:37:26 node2 fenced[2923]: agent "fence_ilo" reports: Unable
> to connect/login to fencing device
> 
> node1 rebooted and get connect to the cluster but now my webby service
> not working see below log :
> 
> Broadcast message from root (Thu Sep 9 14:32:41 2010):
> The system is going down for system halt NOW!
> Sep 9 14:19:22 node1 last message repeated 17 times
> Sep 9 14:32:41 node1 shutdown[25506]: shutting down for system halt
> Sep 9 14:32:41 node1 pcscd: winscard.c:304:SCardConnect() Reader
> E-Gate 0 0 Not Found
> Sep 9 14:32:43 node1 modclusterd: shutdown succeeded
> Sep 9 14:32:43 node1 rgmanager: [25593]: <notice> Shutting down
> Cluster Service Manager...
> Sep 9 14:32:43 node1 clurgmgrd[3457]: <notice> Shutting down
> Sep 9 14:32:43 node1 clurgmgrd[3457]: <notice> Shutting down
> Sep 9 14:32:43 node1 clurgmgrd[3457]: <notice> Stopping service
> service:webby
> Sep 9 14:32:44 node1 avahi-daemon[3378]: Withdrawing address record
> for 192.168.0.30 on eth0.
> Read from remote host node1: Connection reset by peer
> .
> .
> .
> Sep 9 14:35:42 node1 smartd[3585]: Device: /dev/hda, packet devices
> [this device CD/DVD] not SMART capable
> Sep 9 14:35:42 node1 smartd[3585]: Device: /dev/sda, opened
> Sep 9 14:35:42 node1 smartd[3585]: Device: /dev/sda, IE (SMART) not
> enabled, skip device Try 'smartctl -s on /dev/sda' to turn on SMART
> features
> Sep 9 14:35:42 node1 smartd[3585]: Monitoring 0 ATA and 0 SCSI devices
> Sep 9 14:35:42 node1 smartd[3604]: smartd has fork()ed into background
> mode. New PID=3604.
> Sep 9 14:35:42 node1 avahi-daemon[3412]: Service "SFTP File Transfer
> on node1" (/services/sftp-ssh.service) successfully established.
> Sep 9 14:35:45 node1 pcscd: winscard.c:304:SCardConnect() Reader
> E-Gate 0 0 Not Found
> Sep 9 14:35:45 node1 last message repeated 3 times
> Sep 9 14:35:45 node1 kernel: mtrr: type mismatch for d8000000,2000000
> old: uncachable new: write-combining
> Sep 9 14:35:46 node1 clurgmgrd: [3491]: <err> Checking Existence Of
> File /var/run/cluster/apache/apache:httpd.pid [apache:httpd] > Failed
> - File Doesn't Exist
> 
> 
> 
> It seems that there problem in fencing device configuration.
> Please find here my cluster.conf :
> 
> 
> <?xml version="1.0"?>
> <cluster alias="girish" config_version="21" name="girish">
> <fence_daemon clean_start="0" post_fail_delay="0"
> post_join_delay="3"/>
> <clusternodes>
> <clusternode name=" node2.drctmb.com " nodeid="1" votes="1">
> <fence>
> <method name="1">
> <device name="NODE2"/>
> </method>
> </fence>
> </clusternode>
> <clusternode name="node1.drctmb.com" nodeid="2" votes="1">
> <fence>
> <method name="1">
> <device name="NODE1"/>
> </method>
> </fence>
> </clusternode>
> </clusternodes>
> <cman expected_votes="1" two_node="1"/>
> <fencedevices>
> <fencedevice agent="fence_ilo" hostname="node1.drctmb.com"
> login="root" name="NODE1" passwd="redhat123"/>
> <fencedevice agent="fence_ilo" hostname="node2.drctmb.com"
> login="root" name="NODE2" passwd="redhat123"/>
> </fencedevices>
> <rm>
> <failoverdomains>
> <failoverdomain name="prefer_node1" nofailback="0" ordered="1"
> restricted="1">
> <failoverdomainnode name="node2.drctmb.com" priority="2"/>
> <failoverdomainnode name="node1.drctmb.com" priority="1"/>
> </failoverdomain>
> </failoverdomains>
> <resources>
> <fs device="/dev/sda1" force_fsck="0" force_unmount="0" fsid="8669"
> fstype="ext3" mountpoint="/var/www/html" name="docroot"
> self_fence="0"/>
> <ip address="192.168.0.30" monitor_link="1"/>
> <apache config_file="conf/httpd.conf" name="httpd"
> server_root="/etc/httpd" shutdown_wait="5"/>
> </resources>
> <service autostart="1" domain="prefer_node1" exclusive="0"
> name="webby" recovery="relocate">
> <ip ref="192.168.0.30"/>
> <fs ref="docroot"/>
> <apache ref="httpd"/>
> </service>
> </rm>
> <fence_xvmd/>
> </cluster>
> ~
> 
> This is first time am working on Clustering so please help me.
> Appreciate your help.
> 
> Thank you.
> 
> 
> 
> From: Brem Belguebli <brem belguebli gmail com>
> To: linux clustering <linux-cluster redhat com>
> Sent: Thu, September 9, 2010 11:30:28 AM
> Subject: Re: [Linux-cluster] need help - Fencing problem
> 
> try run this from another node of the cluster
> 
> fence_ilo -a "Ilo IP" -l "Ilo user" -p "Ilo passwd" -o reboot
> 
> 
> Additionnally, by connecting thru http to the Ilo, you should be able
> to
> see Ilo logs (in the general tab) and see if it is due to a lack of
> licensing
> 
> 
> On Wed, 2010-09-08 at 22:29 -0700, Girish Prajapati wrote:
> > Hello...
> >
> > I have already configure BIOS for iLO.. but am not sure why i don
> need
> > to shared ??
> > please anybody can help me out for this problem.
> > Do i need any extra setup for fencing device ?
> > thanks
> >
> >
> >
> >
> ______________________________________________________________________
> > From: ESGLinux < esggrupos gmail com >
> > To: linux clustering < linux-cluster redhat com >
> > Sent: Wed, September 8, 2010 2:57:25 PM
> > Subject: Re: [Linux-cluster] need help - Fencing problem
> >
> > Hello,
> >
> >
> > Have you configured the iLO devices entering in the BIOS?
> >
> >
> > I remenber I have to set up the user/pass in the iLO and marked the
> > iLo as not shared
> >
> >
> >
> >
> > HTH,
> >
> >
> > ESG
> >
> > 2010/9/8 Girish Prajapati < girishpati yahoo com >
> > Hello Everybody,
> > i am having problem of fencing a cluster node let me explain
> > indetail :
> > I have installed RHEL 5.4 on HP Prolaint DL280 G5 servers and
> > iLO 2as fencing device. Am managing cluster through Luci -
> > (Conga). itseems everything is working fine. I can reboot
> > cluster nodes through Luci and service get transfer to another
> > node. After rebooting node connect to cluster automatically
> > without any error.
> > Problem is i can not do Fence this node through Luci, when i
> > try to fence any node i get following error :
> >
> > Sep 8 14:51:16 node2 fence_node[9106]: agent "fence_ilo"
> > reports: Unable to connect/login to fencing device
> > Sep 8 14:51:16 node2 fence_node[9106]: Fence of
> > " node1.drctmb.com " was unsuccessful
> >
> > my iLO license is : iLO 2 Advanced Evaluation
> > Do i need to have license of iLO or there is problem in
> > configuration of cluster ?
> > how i can check cluster log in details.
> >
> > Appreciate your help.
> > Thank you in advance.
> >
> > Regards,
> > Girishkumar R Prajapati
> >
> >
> >
> > --
> > Linux-cluster mailing list
> > Linux-cluster redhat com
> > https://www.redhat.com/mailman/listinfo/linux-cluster
> >
> >
> >
> > --
> > Linux-cluster mailing list
> > Linux-cluster redhat com
> > https://www.redhat.com/mailman/listinfo/linux-cluster
> 
> 
> --
> Linux-cluster mailing list
> Linux-cluster redhat com
> https://www.redhat.com/mailman/listinfo/linux-cluster
> 
> 
> 
> --
> Linux-cluster mailing list
> Linux-cluster redhat com
> https://www.redhat.com/mailman/listinfo/linux-cluster
> --
> Linux-cluster mailing list
> Linux-cluster redhat com
> https://www.redhat.com/mailman/listinfo/linux-cluster



------------------------------

--
Linux-cluster mailing list
Linux-cluster redhat com
https://www.redhat.com/mailman/listinfo/linux-cluster

End of Linux-cluster Digest, Vol 77, Issue 5
********************************************
Notice:  This e-mail message, together with any attachments, contains
information of Merck & Co., Inc. (One Merck Drive, Whitehouse Station,
New Jersey, USA 08889), and/or its affiliates Direct contact information
for affiliates is available at 
http://www.merck.com/contact/contacts.html) that may be confidential,
proprietary copyrighted and/or legally privileged. It is intended solely
for the use of the individual or entity named on this message. If you are
not the intended recipient, and have received this message in error,
please notify us immediately by reply e-mail and then delete it from 
your system.



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]