<html> <head> <style> .hmmessage P { margin:0px; padding:0px } body.hmmessage { font-size: 10pt; font-family:Verdana } </style> </head> <body class='hmmessage'> 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 <DIV> ntoughe@hotmail.com</DIV> > From: linux-cluster-request@redhat.com > Subject: Linux-cluster Digest, Vol 64, Issue 16 > To: linux-cluster@redhat.com > Date: Thu, 13 Aug 2009 11:02:36 -0400 > > 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: 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@gmail.com> > Subject: Re: [Linux-cluster] do I have a fence DRAC device? > To: linux clustering <linux-cluster@redhat.com> > Message-ID: > <3128ba140908130427i6ab85406ye6da34073e6a6e97@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@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@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@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@gmail.com> > > => > > > => > On Mon, Aug 10, 2009 at 5:24 AM, ESGLinux<esggrupos@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@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@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/20090813/a4558d27/attachment.html > > ------------------------------ > > Message: 2 > Date: Thu, 13 Aug 2009 14:45:13 +0200 > From: mark benschop <mark.benschop.lists@gmail.com> > Subject: [Linux-cluster] clusterservice stays in 'recovering' state > To: linux-cluster@redhat.com > Message-ID: > <f97c3a70908130545n11ce442ej17d74c9cdc450e45@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@itu.edu.tr> > Subject: Re: [Linux-cluster] Is there any backup heartbeat channel > To: linux-cluster@redhat.com > Message-ID: <20090813161312.11546h2sp6psr814@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@gmail.com den ileti --------- > Tarih: Thu, 13 Aug 2009 10:42:11 +0200 > Kimden: Juan Ramon Martin Blanco <robejrm@gmail.com> > Yanýt Adresi:linux clustering <linux-cluster@redhat.com> > Konu: Re: [Linux-cluster] Is there any backup heartbeat channel > Kime: linux clustering <linux-cluster@redhat.com> > > > > 2009/8/13 Hakan VELIOGLU <veliogluh@itu.edu.tr> > > > >> ----- raju.rajsand@gmail.com den ileti --------- > >> Tarih: Thu, 13 Aug 2009 08:57:15 +0530 > >> Kimden: Rajagopal Swaminathan <raju.rajsand@gmail.com> > >> Yanýt Adresi:linux clustering <linux-cluster@redhat.com> > >> Konu: Re: [Linux-cluster] Is there any backup heartbeat channel > >> Kime: linux clustering <linux-cluster@redhat.com> > >> > >> > >> Greetings, > >>> > >>> 2009/8/12 Hakan VELIOGLU <veliogluh@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@redhat.com > >>> https://www.redhat.com/mailman/listinfo/linux-cluster > >>> > >>> > >> > >> ----- raju.rajsand@gmail.com den iletiyi bitir ----- > >> > >> > >> > >> > >> -- > >> Linux-cluster mailing list > >> Linux-cluster@redhat.com > >> https://www.redhat.com/mailman/listinfo/linux-cluster > >> > > > > > ----- robejrm@gmail.com den iletiyi bitir ----- > > > > > > ------------------------------ > > Message: 4 > Date: Thu, 13 Aug 2009 15:29:43 +0200 > From: Juan Ramon Martin Blanco <robejrm@gmail.com> > Subject: Re: [Linux-cluster] Is there any backup heartbeat channel > To: linux clustering <linux-cluster@redhat.com> > Message-ID: > <8a5668960908130629n6ec05a88n463a3b03da331dae@mail.gmail.com> > Content-Type: text/plain; charset="iso-8859-9" > > 2009/8/13 Hakan VELIOGLU <veliogluh@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@gmail.com den ileti --------- > > Tarih: Thu, 13 Aug 2009 10:42:11 +0200 > > Kimden: Juan Ramon Martin Blanco <robejrm@gmail.com> > > > > Yanýt Adresi:linux clustering <linux-cluster@redhat.com> > > Konu: Re: [Linux-cluster] Is there any backup heartbeat channel > > Kime: linux clustering <linux-cluster@redhat.com> > > > > > > 2009/8/13 Hakan VELIOGLU <veliogluh@itu.edu.tr> > >> > >> ----- raju.rajsand@gmail.com den ileti --------- > >>> Tarih: Thu, 13 Aug 2009 08:57:15 +0530 > >>> Kimden: Rajagopal Swaminathan <raju.rajsand@gmail.com> > >>> Yanýt Adresi:linux clustering <linux-cluster@redhat.com> > >>> Konu: Re: [Linux-cluster] Is there any backup heartbeat channel > >>> Kime: linux clustering <linux-cluster@redhat.com> > >>> > >>> > >>> Greetings, > >>> > >>>> > >>>> 2009/8/12 Hakan VELIOGLU <veliogluh@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@redhat.com > >>>> https://www.redhat.com/mailman/listinfo/linux-cluster > >>>> > >>>> > >>>> > >>> ----- raju.rajsand@gmail.com den iletiyi bitir ----- > >>> > >>> > >>> > >>> > >>> -- > >>> Linux-cluster mailing list > >>> Linux-cluster@redhat.com > >>> https://www.redhat.com/mailman/listinfo/linux-cluster > >>> > >>> > >> > > > > ----- robejrm@gmail.com den iletiyi bitir ----- > > > > > > > > > > -- > > 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/20090813/cd4dc079/attachment.html > > ------------------------------ > > Message: 5 > Date: Thu, 13 Aug 2009 10:07:13 -0400 > From: Nehemias Jahcob <nehemiasjahcob@gmail.com> > Subject: [Linux-cluster] RHCS on KVM > To: linux clustering <linux-cluster@redhat.com> > Message-ID: > <5f61ab380908130707q5c936504k7351d0d6b3459090@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@hotmail.com> > Subject: [Linux-cluster] Cman hang > To: <linux-cluster@redhat.com> > Message-ID: <BAY119-W410E2F250E8B461752CFC9A5050@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@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@redhat.com> > Subject: Re: [Linux-cluster] gfs2 mount hangs > To: Wengang Wang <wen.gang.wang@oracle.com> > Cc: linux clustering <linux-cluster@redhat.com> > Message-ID: <20090813141424.GA8148@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@redhat.com> > Subject: Re: [Linux-cluster] Qdisk question > To: linux clustering <linux-cluster@redhat.com> > Message-ID: <1250174386.23376.1440.camel@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@gmail.com> > Subject: Re: [Linux-cluster] Cman hang > To: linux clustering <linux-cluster@redhat.com> > Message-ID: > <8a5668960908130802p4f5168cbueda86d1e6f1324bb@mail.gmail.com> > Content-Type: text/plain; charset="iso-8859-1" > > On Thu, Aug 13, 2009 at 4:16 PM, NTOUGHE GUY-SERGE <ntoughe@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@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@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@redhat.com > https://www.redhat.com/mailman/listinfo/linux-cluster > > End of Linux-cluster Digest, Vol 64, Issue 16 > ********************************************* <hr />See all the ways you can stay connected <a href='http://www.microsoft.com/windows/windowslive/default.aspx' target='_new'>to friends and family</a></body> </html>