[Linux-cluster] Cman hang
NTOUGHE GUY-SERGE
ntoughe at hotmail.com
Thu Aug 13 15:13:32 UTC 2009
Hi, this is my cluster.conf
<?xml version="1.0"?>
<cluster alias="arevclust" config_version="21" name="arevclust">
<fence_daemon clean_start="0" post_fail_delay="0" post_join_delay="3"/>
<clusternodes>
<clusternode name="host1" nodeid="1" votes="1">
<fence>
<method name="2">
<device name=""/>
</method>
</fence>
<multicast addr="" interface=""/>
</clusternode>
<clusternode name="host2" nodeid="2" votes="1">
<fence>
<method name="1">
<device name=""/>
</method>
<method name=""/>
</fence>
<multicast addr="" interface=""/>
</clusternode>
</clusternodes>
<cman expected_votes="" two_node="">
<multicast addr=""/>
</cman>
<fencedevices>
<fencedevice agent="fence_brocade" ipaddr="" login="" name="" passwd=""/>
</fencedevices>
<rm>
<failoverdomains>
</failoverdomains>
<resources>
</resources>
</rm>
</cluster>
Regards
ntoughe at hotmail.com
> From: linux-cluster-request at redhat.com
> Subject: Linux-cluster Digest, Vol 64, Issue 16
> To: linux-cluster at redhat.com
> Date: Thu, 13 Aug 2009 11:02:36 -0400
>
> Send Linux-cluster mailing list submissions to
> linux-cluster at 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 at redhat.com
>
> You can reach the person managing the list at
> linux-cluster-owner at 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: do I have a fence DRAC device? (ESGLinux)
> 2. clusterservice stays in 'recovering' state (mark benschop)
> 3. Re: Is there any backup heartbeat channel (Hakan VELIOGLU)
> 4. Re: Is there any backup heartbeat channel
> (Juan Ramon Martin Blanco)
> 5. RHCS on KVM (Nehemias Jahcob)
> 6. Cman hang (NTOUGHE GUY-SERGE)
> 7. Re: gfs2 mount hangs (David Teigland)
> 8. Re: Qdisk question (Lon Hohberger)
> 9. Re: Cman hang (Juan Ramon Martin Blanco)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Thu, 13 Aug 2009 13:27:16 +0200
> From: ESGLinux <esggrupos at gmail.com>
> Subject: Re: [Linux-cluster] do I have a fence DRAC device?
> To: linux clustering <linux-cluster at redhat.com>
> Message-ID:
> <3128ba140908130427i6ab85406ye6da34073e6a6e97 at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hi,
> I couldn´t reboot my system yet but I have installed the openmanage
> packages:
>
> srvadmin-omacore-5.4.0-260
> srvadmin-iws-5.4.0-260
> srvadmin-syscheck-5.4.0-260
> srvadmin-rac5-components-5.4.0-260
> srvadmin-deng-5.4.0-260
> srvadmin-ipmi-5.4.0-260.DUP
> srvadmin-racadm5-5.4.0-260
> srvadmin-omauth-5.4.0-260.rhel5
> srvadmin-hapi-5.4.0-260
> srvadmin-cm-5.4.0-260
> srvadmin-racdrsc5-5.4.0-260
> srvadmin-omilcore-5.4.0-260
> srvadmin-isvc-5.4.0-260
> srvadmin-storage-5.4.0-260
> srvadmin-jre-5.4.0-260
> srvadmin-omhip-5.4.0-260
>
> Now I have the command racadm but when I try to execut it I get this:
>
> racadm config -g cfgSerial -o cfgSerialTelnetEnable 1
> ERROR: RACADM is unable to process the requested subcommand because there is
> no
> local RAC configuration to communicate with.
>
> Local RACADM subcommand execution requires the following:
>
> 1. A Remote Access Controller (RAC) must be present on the managed server
> 2. Appropriate managed node software must be installed and running on the
> server
>
>
> What do I need to install/start? or until I configure the bios I can´t get
> this work?
>
> Greetings
>
> ESG
>
>
> 2009/8/11 <bergman at merctech.com>
>
> >
> >
> > In the message dated: Tue, 11 Aug 2009 14:14:03 +0200,
> > The pithy ruminations from Juan Ramon Martin Blanco on
> > <Re: [Linux-cluster] do I have a fence DRAC device?> were:
> > => --===============1917368601==
> > => Content-Type: multipart/alternative;
> > boundary=0016364c7c07663f600470dca3b8
> > =>
> > => --0016364c7c07663f600470dca3b8
> > => Content-Type: text/plain; charset=ISO-8859-1
> > => Content-Transfer-Encoding: quoted-printable
> > =>
> > => On Tue, Aug 11, 2009 at 2:03 PM, ESGLinux <esggrupos at gmail.com> wrote:
> > =>
> > => > Thanks
> > => > I=B4ll check it when I could reboot the server.
> > => >
> > => > greetings,
> > => >
> > => You have a BMC ipmi in the first network interface, it can be configured
> > at
> > => boot time (I don't remember if inside the BIOS or pressing
> > cntrl+something
> > => during boot)
> > =>
> >
> > Based on my notes, here's how I configured the DRAC interface on a Dell
> > 1950
> > for use as a fence device:
> >
> > Configuring the card from Linux depending on the installation of
> > Dell's
> > OMSA package. Once that's installed, use the following
> > commands:
> >
> > racadm config -g cfgSerial -o cfgSerialTelnetEnable 1
> > racadm config -g cfgLanNetworking -o cfgDNSRacName
> > HOSTNAME_FOR_INTERFACE
> > racadm config -g cfgDNSDomainName DOMAINNAME_FOR_INTERFACE
> > racadm config -g cfgUserAdmin -o cfgUserAdminPassword -i 2
> > PASSWORD
> > racadm config -g cfgNicEnable 1
> > racadm config -g cfgNicIpAddress WWW.XXX.YYY.ZZZ
> > racadm config -g cfgNicNetmask WWW.XXX.YYY.ZZZ
> > racadm config -g cfgNicGateway WWW.XXX.YYY.ZZZ
> > racadm config -g cfgNicUseDhcp 0
> >
> >
> > I also save a backup of the configuration with:
> >
> > racadm getconfig -f ~/drac_config
> >
> >
> > Hope this helps,
> >
> > Mark
> >
> > ----
> > Mark Bergman voice: 215-662-7310
> > mark.bergman at uphs.upenn.edu fax: 215-614-0266
> > System Administrator Section of Biomedical Image Analysis
> > Department of Radiology University of Pennsylvania
> > PGP Key: https://www.rad.upenn.edu/sbia/bergman
> >
> >
> > => Greetings,
> > => Juanra
> > =>
> > => >
> > => > ESG
> > => >
> > => > 2009/8/10 Paras pradhan <pradhanparas at gmail.com>
> > => >
> > => > On Mon, Aug 10, 2009 at 5:24 AM, ESGLinux<esggrupos at gmail.com> wrote:
> > => >> > Hi all,
> > => >> > I was designing a 2 node cluster and I was going to use 2 servers
> > DELL
> > => >> > PowerEdge 1950. I was going to buy a DRAC card to use for fencing
> > but
> > => >> > running several commands in the servers I have noticed that when I
> > run
> > => >> this
> > => >> > command:
> > => >> > #ipmitool lan print
> > => >> > Set in Progress : Set Complete
> > => >> > Auth Type Support : NONE MD2 MD5 PASSWORD
> > => >> > Auth Type Enable : Callback : MD2 MD5
> > => >> > : User : MD2 MD5
> > => >> > : Operator : MD2 MD5
> > => >> > : Admin : MD2 MD5
> > => >> > : OEM : MD2 MD5
> > => >> > IP Address Source : Static Address
> > => >> > IP Address : 0.0.0.0
> > => >> > Subnet Mask : 0.0.0.0
> > => >> > MAC Address : 00:1e:c9:ae:6f:7e
> > => >> > SNMP Community String : public
> > => >> > IP Header : TTL=0x40 Flags=0x40 Precedence=0x00 TOS=0x10
> > => >> > Default Gateway IP : 0.0.0.0
> > => >> > Default Gateway MAC : 00:00:00:00:00:00
> > => >> > Backup Gateway IP : 0.0.0.0
> > => >> > Backup Gateway MAC : 00:00:00:00:00:00
> > => >> > 802.1q VLAN ID : Disabled
> > => >> > 802.1q VLAN Priority : 0
> > => >> > RMCP+ Cipher Suites : 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14
> > => >> > Cipher Suite Priv Max : aaaaaaaaaaaaaaa
> > => >> > : X=Cipher Suite Unused
> > => >> > : c=CALLBACK
> > => >> > : u=USER
> > => >> > : o=OPERATOR
> > => >> > : a=ADMIN
> > => >> > : O=OEM
> > => >> > does this mean that I already have an ipmi card (not configured)
> > that
> > => I
> > => >> can
> > => >> > use for fencing? if the anwser is yes, where hell must I configure
> > it?
> > => I
> > => >> > don=B4t see wher can I do it.
> > => >> > If I haven=B4t a fencing device which one do you recommed to use?
> > => >> > Thanks in advance
> > => >> > ESG
> > => >> >
> > => >> > --
> > => >> > Linux-cluster mailing list
> > => >> > Linux-cluster at redhat.com
> > => >> > https://www.redhat.com/mailman/listinfo/linux-cluster
> > => >> >
> > => >>
> > => >> Yes you have IPMI and if you are using 1950 Dell, DRAC should be
> > there
> > => >> too. You can see if you have DRAC or not when the server starts and
> > => >> before the loading of the OS.
> > => >>
> > => >> I have 1850s and I am using DRAC for fencing.
> > => >>
> > => >>
> > => >> Paras.
> > => >>
> > => >> --
> > => >> Linux-cluster mailing list
> > => >> Linux-cluster at redhat.com
> > => >> https://www.redhat.com/mailman/listinfo/linux-cluster
> > => >>
> > => >
> > => >
> >
> >
> >
> > --
> > Linux-cluster mailing list
> > Linux-cluster at 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/20090813/a4558d27/attachment.html
>
> ------------------------------
>
> Message: 2
> Date: Thu, 13 Aug 2009 14:45:13 +0200
> From: mark benschop <mark.benschop.lists at gmail.com>
> Subject: [Linux-cluster] clusterservice stays in 'recovering' state
> To: linux-cluster at redhat.com
> Message-ID:
> <f97c3a70908130545n11ce442ej17d74c9cdc450e45 at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hi All,
>
> I've a problem with a clusterservice. The service was started up while one
> of the resources, an NFS, export was not accessible.
> Therefore the service never started up right but got into the 'recovering'
> state.
> In the mean time the NFS exports are setup properly but to no avail.
> Stopping the clusterservice, using clusvcadm -d <service>, will result in
> the service going down but staying in the 'recovering' state.
> Starting it again doesn't work. The service doesn't start and stays in the
> recovery status.
> I'm suspecting rgmanager lost it somehow.
>
> Anybody had any ideas on what could be the problem and how to resolve it ?
>
> Thanks in advance,
> Mark
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: https://www.redhat.com/archives/linux-cluster/attachments/20090813/31731cd3/attachment.html
>
> ------------------------------
>
> Message: 3
> Date: Thu, 13 Aug 2009 16:13:12 +0300
> From: Hakan VELIOGLU <veliogluh at itu.edu.tr>
> Subject: Re: [Linux-cluster] Is there any backup heartbeat channel
> To: linux-cluster at redhat.com
> Message-ID: <20090813161312.11546h2sp6psr814 at webmail.itu.edu.tr>
> Content-Type: text/plain; charset=ISO-8859-9; DelSp="Yes";
> format="flowed"
>
> Thanks for all the answers.
>
> I think there is realy no backup heartbeat channel. Maybe the reason
> is GFS. DLM works on the heartbeat channel. If you lost your heartbeat
> you lose your lock consistency so it is better to fence the other
> node. For this reason I think if you don't have enough network
> interface on server and switch, loosing the heartbeat network may shut
> all the cluster members.
>
> Hakan VELÝOÐLU
>
>
> ----- robejrm at gmail.com den ileti ---------
> Tarih: Thu, 13 Aug 2009 10:42:11 +0200
> Kimden: Juan Ramon Martin Blanco <robejrm at gmail.com>
> Yanýt Adresi:linux clustering <linux-cluster at redhat.com>
> Konu: Re: [Linux-cluster] Is there any backup heartbeat channel
> Kime: linux clustering <linux-cluster at redhat.com>
>
>
> > 2009/8/13 Hakan VELIOGLU <veliogluh at itu.edu.tr>
> >
> >> ----- raju.rajsand at gmail.com den ileti ---------
> >> Tarih: Thu, 13 Aug 2009 08:57:15 +0530
> >> Kimden: Rajagopal Swaminathan <raju.rajsand at gmail.com>
> >> Yanýt Adresi:linux clustering <linux-cluster at redhat.com>
> >> Konu: Re: [Linux-cluster] Is there any backup heartbeat channel
> >> Kime: linux clustering <linux-cluster at redhat.com>
> >>
> >>
> >> Greetings,
> >>>
> >>> 2009/8/12 Hakan VELIOGLU <veliogluh at itu.edu.tr>:
> >>>
> >>>> Hi list,
> >>>>
> >>>> I am trying a two node cluster with RH 5.3 on Sun X4150 hardware. I use a
> >>>>
> >>>
> >>> IIRC, Sun x4150 has four ethernet ports. Two can be used for outside
> >>> networking and two can be bonded and used for heartbeat.
> >>>
> >> I think, I couldn't explain my networking. I use two ethernet ports for xen
> >> vm which are trunk and bonded ports. Then there left two. Our network
> >> topology (which is out of my control) available for one port for server
> >> control (SSH).
> >
> > So you can't use a bonded port for both server management and cluster
> > communications, can you? You can configure an active-passive bonding and
> > then you can have many virtual interfaces on top of that, i.e: bond0:0,
> > bond0:1 and assign them the ip addesses you need.
> >
> >
> > I use the other one with a cross over cable for heartbeat. So there is no
> >> way for bonding these two interfaces. Of course if I buy an extra switch I
> >> may do this.
> >
> > You can connect them to the same switch (though you lost kind of
> > redundancy), or you can use two crossover cables and move the management IP
> > to the same ports you are using for the vm's.
> >
> > Greetings,
> > Juanra
> >
> >>
> >> I don't realy understand why there is no backup heartbeat channel. LVS and
> >> MS cluster has this ability.
> >>
> >>>
> >>> ALOM can be used for fencing and can be on a seperate subnet if required.
> >>>
> >> I used this for fencing_ipmilan.
> >>
> >>>
> >>> Regards
> >>>
> >>> Rajagopal
> >>>
> >>> --
> >>> Linux-cluster mailing list
> >>> Linux-cluster at redhat.com
> >>> https://www.redhat.com/mailman/listinfo/linux-cluster
> >>>
> >>>
> >>
> >> ----- raju.rajsand at gmail.com den iletiyi bitir -----
> >>
> >>
> >>
> >>
> >> --
> >> Linux-cluster mailing list
> >> Linux-cluster at redhat.com
> >> https://www.redhat.com/mailman/listinfo/linux-cluster
> >>
> >
>
>
> ----- robejrm at gmail.com den iletiyi bitir -----
>
>
>
>
>
> ------------------------------
>
> Message: 4
> Date: Thu, 13 Aug 2009 15:29:43 +0200
> From: Juan Ramon Martin Blanco <robejrm at gmail.com>
> Subject: Re: [Linux-cluster] Is there any backup heartbeat channel
> To: linux clustering <linux-cluster at redhat.com>
> Message-ID:
> <8a5668960908130629n6ec05a88n463a3b03da331dae at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-9"
>
> 2009/8/13 Hakan VELIOGLU <veliogluh at itu.edu.tr>
>
> > Thanks for all the answers.
> >
> > I think there is realy no backup heartbeat channel. Maybe the reason is
> > GFS. DLM works on the heartbeat channel. If you lost your heartbeat you lose
> > your lock consistency so it is better to fence the other node. For this
> > reason I think if you don't have enough network interface on server and
> > switch, loosing the heartbeat network may shut all the cluster members.
> >
> There is no backup heartbeat channel because you should do the backup at a
> operating system level, i.e: bonding
> That's why you should use a bonded interface for the heartbeat channel with
> at least 2 ethernet slaves; going further (for better redundancy) each of
> the slaves should be on a different network card and you should connect the
> each slave to a different switch.
> But what I am trying to explain, is that you can use that bonded logical
> interface also for things different from hearbeat. ;)
>
> Greetings,
> Juanra
>
>
> > Hakan VELÝOÐLU
> >
> >
> > ----- robejrm at gmail.com den ileti ---------
> > Tarih: Thu, 13 Aug 2009 10:42:11 +0200
> > Kimden: Juan Ramon Martin Blanco <robejrm at gmail.com>
> >
> > Yanýt Adresi:linux clustering <linux-cluster at redhat.com>
> > Konu: Re: [Linux-cluster] Is there any backup heartbeat channel
> > Kime: linux clustering <linux-cluster at redhat.com>
> >
> >
> > 2009/8/13 Hakan VELIOGLU <veliogluh at itu.edu.tr>
> >>
> >> ----- raju.rajsand at gmail.com den ileti ---------
> >>> Tarih: Thu, 13 Aug 2009 08:57:15 +0530
> >>> Kimden: Rajagopal Swaminathan <raju.rajsand at gmail.com>
> >>> Yanýt Adresi:linux clustering <linux-cluster at redhat.com>
> >>> Konu: Re: [Linux-cluster] Is there any backup heartbeat channel
> >>> Kime: linux clustering <linux-cluster at redhat.com>
> >>>
> >>>
> >>> Greetings,
> >>>
> >>>>
> >>>> 2009/8/12 Hakan VELIOGLU <veliogluh at itu.edu.tr>:
> >>>>
> >>>> Hi list,
> >>>>>
> >>>>> I am trying a two node cluster with RH 5.3 on Sun X4150 hardware. I use
> >>>>> a
> >>>>>
> >>>>>
> >>>> IIRC, Sun x4150 has four ethernet ports. Two can be used for outside
> >>>> networking and two can be bonded and used for heartbeat.
> >>>>
> >>>> I think, I couldn't explain my networking. I use two ethernet ports for
> >>> xen
> >>> vm which are trunk and bonded ports. Then there left two. Our network
> >>> topology (which is out of my control) available for one port for server
> >>> control (SSH).
> >>>
> >>
> >> So you can't use a bonded port for both server management and cluster
> >> communications, can you? You can configure an active-passive bonding and
> >> then you can have many virtual interfaces on top of that, i.e: bond0:0,
> >> bond0:1 and assign them the ip addesses you need.
> >>
> >>
> >> I use the other one with a cross over cable for heartbeat. So there is no
> >>
> >>> way for bonding these two interfaces. Of course if I buy an extra switch
> >>> I
> >>> may do this.
> >>>
> >>
> >> You can connect them to the same switch (though you lost kind of
> >> redundancy), or you can use two crossover cables and move the management
> >> IP
> >> to the same ports you are using for the vm's.
> >>
> >> Greetings,
> >> Juanra
> >>
> >>
> >>> I don't realy understand why there is no backup heartbeat channel. LVS
> >>> and
> >>> MS cluster has this ability.
> >>>
> >>>
> >>>> ALOM can be used for fencing and can be on a seperate subnet if
> >>>> required.
> >>>>
> >>>> I used this for fencing_ipmilan.
> >>>
> >>>
> >>>> Regards
> >>>>
> >>>> Rajagopal
> >>>>
> >>>> --
> >>>> Linux-cluster mailing list
> >>>> Linux-cluster at redhat.com
> >>>> https://www.redhat.com/mailman/listinfo/linux-cluster
> >>>>
> >>>>
> >>>>
> >>> ----- raju.rajsand at gmail.com den iletiyi bitir -----
> >>>
> >>>
> >>>
> >>>
> >>> --
> >>> Linux-cluster mailing list
> >>> Linux-cluster at redhat.com
> >>> https://www.redhat.com/mailman/listinfo/linux-cluster
> >>>
> >>>
> >>
> >
> > ----- robejrm at gmail.com den iletiyi bitir -----
> >
> >
> >
> >
> > --
> > Linux-cluster mailing list
> > Linux-cluster at 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/20090813/cd4dc079/attachment.html
>
> ------------------------------
>
> Message: 5
> Date: Thu, 13 Aug 2009 10:07:13 -0400
> From: Nehemias Jahcob <nehemiasjahcob at gmail.com>
> Subject: [Linux-cluster] RHCS on KVM
> To: linux clustering <linux-cluster at redhat.com>
> Message-ID:
> <5f61ab380908130707q5c936504k7351d0d6b3459090 at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hi.
>
> How to create a cluster of 2 nodes in rhel5.4 (or Fedora 10) with KVM?
>
> With XEN follow this guide:
> http://sources.redhat.com/cluster/wiki/VMClusterCookbook?highlight =
> (CategoryHowTo).
>
> Do you have a guide to implementation of RHCS in KVM?
>
> Thank you all.
> NJ
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: https://www.redhat.com/archives/linux-cluster/attachments/20090813/f3a69a80/attachment.html
>
> ------------------------------
>
> Message: 6
> Date: Thu, 13 Aug 2009 14:16:47 +0000
> From: NTOUGHE GUY-SERGE <ntoughe at hotmail.com>
> Subject: [Linux-cluster] Cman hang
> To: <linux-cluster at redhat.com>
> Message-ID: <BAY119-W410E2F250E8B461752CFC9A5050 at phx.gbl>
> Content-Type: text/plain; charset="iso-8859-1"
>
>
>
> Hi gurus,
>
> i installed RHEL 5.3 on 2 servers which participating to a cluster composed of these 2 nodes:
> kernel version:
> kernel-headers-2.6.18-128.el5
> kernel-devel-2.6.18-128.el5
> kernel-2.6.18-128.el5
> cman-devel-2.0.98-1.el5_3.1
> cman-2.0.98-1.el5_3.1
> cluster-cim-0.12.1-2.el5
> lvm2-cluster-2.02.40-7.el5
> cluster-snmp-0.12.1-2.el5
> modcluster-0.12.1-2.el5
> When i want to start cman the following message is sent:
> cman not started: Multicast and node address families differ. /usr/sbin/cman_tool: aisexec daemon didn't start
> [FAILED]
>
> I trier to mount gfs2
> and i got theses messages:
> # mount -t gfs2 /dev/VolGroup01/LogVol01 /appli/prod --o lockTablename=arvclust:/appli/prod, Lockproto=lock_dlm
>
> /sbin/mount.gfs2: can't connect to gfs_controld: Connection refused
>
> /sbin/mount.gfs2: can't connect to gfs_controld: Connection refused
>
> /sbin/mount.gfs2: can't connect to gfs_controld: Connection refused
>
> /sbin/mount.gfs2: can't connect to gfs_controld: Connection refused
>
> do you have any clues?
> Please it's an hurry, i waste long time to lok for solution help
> regards
>
>
>
>
>
>
>
>
> ntoughe at hotmail.com
>
>
> _________________________________________________________________
> With Windows Live, you can organize, edit, and share your photos.
> http://www.microsoft.com/middleeast/windows/windowslive/products/photo-gallery-edit.aspx
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: https://www.redhat.com/archives/linux-cluster/attachments/20090813/0a55101d/attachment.html
>
> ------------------------------
>
> Message: 7
> Date: Thu, 13 Aug 2009 09:14:24 -0500
> From: David Teigland <teigland at redhat.com>
> Subject: Re: [Linux-cluster] gfs2 mount hangs
> To: Wengang Wang <wen.gang.wang at oracle.com>
> Cc: linux clustering <linux-cluster at redhat.com>
> Message-ID: <20090813141424.GA8148 at redhat.com>
> Content-Type: text/plain; charset=us-ascii
>
> On Thu, Aug 13, 2009 at 02:22:11PM +0800, Wengang Wang wrote:
> > <cman two_node="1" expected_votes="2"/>
>
> That's not a valid combination, two_node="1" requires expected_votes="1".
>
> You didn't mention which userspace cluster version/release you're using, or
> include any status about the cluster. Before trying to mount gfs on either
> node, collect from both nodes:
>
> cman_tool status
> cman_tool nodes
> group_tool
>
> Then mount on the first node and collect the same information, then try
> mounting on the second node, collect the same information, and look for any
> errors in /var/log/messages.
>
> Since you're using new kernels, you need to be using the cluster 3.0 userspace
> code. You're using the old manual fencing config. There is no more
> fence_manual; the new way to configure manual fencing is to not configure any
> fencing at all. So, your cluster.conf should look like this:
>
> <?xml version="1.0"?>
> <cluster name="testgfs2" config_version="1">
> <cman two_node="1" expected_votes="1"/>
> <clusternodes>
> <clusternode name="cool" nodeid="1"/>
> <clusternode name="desk" nodeid="2"/>
> </clusternodes>
> </cluster>
>
> Dave
>
>
>
> ------------------------------
>
> Message: 8
> Date: Thu, 13 Aug 2009 10:39:46 -0400
> From: Lon Hohberger <lhh at redhat.com>
> Subject: Re: [Linux-cluster] Qdisk question
> To: linux clustering <linux-cluster at redhat.com>
> Message-ID: <1250174386.23376.1440.camel at localhost.localdomain>
> Content-Type: text/plain
>
> On Thu, 2009-08-13 at 00:45 +0200, brem belguebli wrote:
>
> > My understanding of qdisk is that it is used as a tie-breaker, but it
> > looks like it is more a heatbeat vector than a simple tie-breaker.
>
> Right, it's a secondary membership algorithm.
>
>
> > Until here, no real problem indeed, if the site gets apart from the
> > other prod site and also from the third site (hosting the iscsi target
> > qdisk) the 2 nodes from the failing site get evicted from the cluster.
> >
> >
> > But, what if my third site gets isolated while the 2 prod ones are
> > fine ?
>
> Qdisk votes will not be presented to CMAN any more, but the two sites
> should remain online if they still have a "majority" of votes.
>
>
> > The real question is what happens in case all the nodes loose access
> > to the qdisk while they're still able to see each others ?
>
> Qdisk is just a vote like other voting mechanisms. If all nodes lose
> access at the same time, it should behave like a node death. However,
> the default action if _one_ node loses access is to kill that node (even
> if CMAN still sees it).
>
>
> > The 4 nodes have each 1 vote and the qdisk 1 vote. The expected quorum
> > is 3.
>
>
> > If I loose the qdisk, the number of votes falls to 4, the cluster is
> > quorate (4>3) but it looks like everything goes bad, each node
> > deactivate itself as it can't write its alive status (--> heartbeat
> > vector) to the qdisk even if the network heartbeating is working
> > fine.
>
> What happens specifically? Most of the actions qdiskd performs are
> configurable. For example, if the nodes are rebooting, you can turn
> that behavior off.
>
>
>
> I wrote a simple 'ping' tiebreaker based the behaviors in RHEL3. It
> functions in many ways in the same manner as qdiskd with respect to vote
> advertisement to CMAN, but without needing a disk - maybe you would find
> it useful?
>
> http://people.redhat.com/lhh/qnet.tar.gz
>
> -- Lon
>
>
>
> ------------------------------
>
> Message: 9
> Date: Thu, 13 Aug 2009 17:02:15 +0200
> From: Juan Ramon Martin Blanco <robejrm at gmail.com>
> Subject: Re: [Linux-cluster] Cman hang
> To: linux clustering <linux-cluster at redhat.com>
> Message-ID:
> <8a5668960908130802p4f5168cbueda86d1e6f1324bb at mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> On Thu, Aug 13, 2009 at 4:16 PM, NTOUGHE GUY-SERGE <ntoughe at hotmail.com>wrote:
>
> >
> > Hi gurus,
> >
> > i installed RHEL 5.3 on 2 servers which participating to a cluster
> > composed of these 2 nodes:
> > kernel version:
> > kernel-headers-2.6.18-128.el5
> > kernel-devel-2.6.18-128.el5
> > kernel-2.6.18-128.el5
> > cman-devel-2.0.98-1.el5_3.1
> > cman-2.0.98-1.el5_3.1
> > cluster-cim-0.12.1-2.el5
> > lvm2-cluster-2.02.40-7.el5
> > cluster-snmp-0.12.1-2.el5
> > modcluster-0.12.1-2.el5
> > When i want to start cman the following message is sent:
> > cman not started: Multicast and node address families differ.
> > /usr/sbin/cman_tool: aisexec daemon didn't start
> > [FAILED]
> >
> Please, show us your cluster.conf file so we can help.
>
> Regards,
> Juanra
>
> >
> > I trier to mount gfs2
> > and i got theses messages:
> > # mount -t gfs2 /dev/VolGroup01/LogVol01 /appli/prod --o
> > lockTablename=arvclust:/appli/prod, Lockproto=lock_dlm
> >
> > /sbin/mount.gfs2: can't connect to gfs_controld: Connection refused
> >
> > /sbin/mount.gfs2: can't connect to gfs_controld: Connection refused
> >
> > /sbin/mount.gfs2: can't connect to gfs_controld: Connection refused
> >
> > /sbin/mount.gfs2: can't connect to gfs_controld: Connection refused
> >
> > do you have any clues?
> > Please it's an hurry, i waste long time to lok for solution help
> > regards
> >
> >
> >
> >
> >
> >
> >
> >
> > ntoughe at hotmail.com
> >
> >
> >
> > ------------------------------
> > With Windows Live, you can organize, edit, and share your photos.<http://www.microsoft.com/middleeast/windows/windowslive/products/photo-gallery-edit.aspx>
> >
> > --
> > Linux-cluster mailing list
> > Linux-cluster at 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/20090813/9ecbcab1/attachment.html
>
> ------------------------------
>
> --
> Linux-cluster mailing list
> Linux-cluster at redhat.com
> https://www.redhat.com/mailman/listinfo/linux-cluster
>
> End of Linux-cluster Digest, Vol 64, Issue 16
> *********************************************
_________________________________________________________________
Show them the way! Add maps and directions to your party invites.
http://www.microsoft.com/windows/windowslive/products/events.aspx
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/linux-cluster/attachments/20090813/45029caa/attachment.htm>
More information about the Linux-cluster
mailing list