Restart LVM with out dismounting LUN/FS. Here is what u have to do. #killall clvmd #/usr/sbin/clvmd
-----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of [email protected] Sent: Thursday, January 15, 2009 12:00 PM To: [email protected] Subject: Linux-cluster Digest, Vol 57, Issue 14 Send Linux-cluster mailing list submissions to [email protected] 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 [email protected] You can reach the person managing the list at [email protected] When replying, please edit your Subject line so it is more specific than "Re: Contents of Linux-cluster digest..." Today's Topics: 1. Create Logical Volume (Cluster Management) 2. Re: Re: Fencing test (Paras pradhan) 3. Documentation enhancement requests/patches? (denis) 4. Re: List Cluster Resources (denis) 5. "Simple" Managed NFS setup (denis) 6. cman-2.0.98-1.el5 / question about a problem when launching cman (Alain.Moulle) 7. Re: Documentation enhancement requests/patches? (Bob Peterson) 8. Re: cman-2.0.98-1.el5 / question about a problem when launching cman (Chrissie Caulfield) 9. Re: [Openais] cman in RHEL 5 cluster suite and Openais (Chrissie Caulfield) 10. GFS/clvmd question (Gary Romo) ---------------------------------------------------------------------- Message: 1 Date: Wed, 14 Jan 2009 19:29:23 +0100 From: "Cluster Management" <[email protected]> Subject: [Linux-cluster] Create Logical Volume To: <[email protected]> Message-ID: <008201c97676$05c36600$114a32...@it> Content-Type: text/plain; charset="us-ascii" Hi all, i have a two_node cluster RHEL 5 and an external ISCSI storage. I use XEN for virtualizzation purpose. When i create a new LUN in my storage i use hot_add command to discover it from nodes. The problem is that i have to restart clvmd to be able to create a new Logical Volume. This operation is very critical because i have to stop or migrate each VM running on the node ed i have to umout their own LUN. Is there a way to update clvmd without restarting? Thanks a lot, -- Francesco Gallo XiNet S.r.L. gallo (at) xinet (dot) it -------------- next part -------------- An HTML attachment was scrubbed... URL: https://www.redhat.com/archives/linux-cluster/attachments/20090114/e827f 01f/attachment.html ------------------------------ Message: 2 Date: Wed, 14 Jan 2009 13:48:58 -0600 From: "Paras pradhan" <[email protected]> Subject: Re: [Linux-cluster] Re: Fencing test To: "linux clustering" <[email protected]> Message-ID: <[email protected]> Content-Type: text/plain; charset=ISO-8859-1 On Thu, Jan 8, 2009 at 10:57 PM, Rajagopal Swaminathan <[email protected]> wrote: > Greetings, > > On Fri, Jan 9, 2009 at 12:09 AM, Paras pradhan <[email protected]> wrote: >> >> >> In an act to solve my fencing issue in my 2 node cluster, i tried to >> run fence_ipmi to check if fencing is working or not. I need to know >> what is my problem >> >> - >> [r...@ha1lx ~]# fence_ipmilan -a 10.42.21.28 -o off -l admin -p admin >> Powering off machine @ IPMI:10.42.21.28...ipmilan: Failed to connect >> after 30 seconds >> Failed >> [r...@ha1lx ~]# >> --------------- >> >> >> Here 10.42.21.28 is an IP address assigned to IPMI interface and I am >> running this command in the same host. >> > > Sorry couldn't respond earlier as I do this on personal time (which as > useual limited for us IT guys and gals ;-) ) and not during work per > se.. > > Do not run fence script from the node that you want to fence. > > Let us say you want to fence node 3. > 1. Try pinging the node 3's IPMI from node 4. I should be successful > 2. Issue the fence command from Node 4 with IP of Node 3 IPMI as argument . > > > HTH > > With warm regards > > Rajagopal > > -- > Linux-cluster mailing list > [email protected] > https://www.redhat.com/mailman/listinfo/linux-cluster > Yes as you said, I am able to power down node4 using node3, so it seems ipmi is working fine. But I dunno what is going on with my two node cluster. Can a red hat cluster operates fine in a two nodes mode? Do i need qdisk or it is optional. Which area do i need to focus to run my 2 nodes red hat cluster using ipmi as fencing device. Thanks Paras. ------------------------------ Message: 3 Date: Thu, 15 Jan 2009 10:49:11 +0100 From: denis <[email protected]> Subject: [Linux-cluster] Documentation enhancement requests/patches? To: [email protected] Message-ID: <[email protected]> Content-Type: text/plain; charset=ISO-8859-1 Hi, After getting to know the "Configuring and Managing a Red Hat Cluster" documentation [1] fairly well, I have a few enhancement suggestions. What is the best way to submit these? [1] http://www.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/5.2/html/Clust er_Administration/ Regards -- Denis Braekhus ------------------------------ Message: 4 Date: Thu, 15 Jan 2009 10:52:00 +0100 From: denis <[email protected]> Subject: [Linux-cluster] Re: List Cluster Resources To: [email protected] Message-ID: <[email protected]> Content-Type: text/plain; charset=ISO-8859-1 Chaitanya Kulkarni wrote: > Hi All, > > I am new to the RHEL Clusters. Is there any way, (other than the > cluster.conf file) using which we can view / list all the Cluster > Resources that are used under the Cluster Service (Resource Group)? Some > command which might give some output as - > > Service Name = Service1 > > Resources - > IP Address = <Value> > File System = <Value> > Script = <Value> Hi Chaitanya, I recently discovered the rg_test tool, it might be of help to you. It does currently not have a man page, but check the "Configuring and Managing a Red Hat Cluster" chapter "Debugging and Testing Services and Resource Ordering" [1] for usage. Hope this is of some help to you. [1] http://www.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/5.2/html/Clust er_Administration/s1-clust-rsc-testing-config-CA.html Regards -- Denis Braekhus ------------------------------ Message: 5 Date: Thu, 15 Jan 2009 12:00:03 +0100 From: denis <[email protected]> Subject: [Linux-cluster] "Simple" Managed NFS setup To: [email protected] Message-ID: <[email protected]> Content-Type: text/plain; charset=ISO-8859-1 Hi, I have begun a setup with a pretty simple 3-node cluster and a couple of services. One of these is NFS, and I have setup the basics as laid out in the included cluster.conf below. A couple of questions : 1. Do I need to keep the nfs-state information on the NFS_homes volume so as to keep it in sync between clusternodes? 2. The nfsclient name="nfs" is added to enable the current NFS serving node to mount its own export, otherwise I got Jan 15 11:41:03 node03 mountd[14229]: mount request from unknown host XX.XX.XX.174 for /mnt/nfshome (/mnt/nfshome) This is obviously caused by the mount connecting as the NFS service address instead of the hostaddress, what is the best way to resolve this? Mounting with the serviceaddress is not a good solution it seems, as failing the service over is problematic when that address is in use locally. 3. I read "The Red Hat Cluster Suite NFS Cookbook" [1], as the reference Red Hat documentation was a bit thin regarding best practices. Is there more documentation available to read? Any tips/pointers/help highly appreciated. <rm> <failoverdomains> <failoverdomain name="failover_nfshome" ordered="1" restricted="1"> <failoverdomainnode name="node01.domain" priority="30"/> <failoverdomainnode name="node02.domain" priority="30"/> <failoverdomainnode name="node03.domain" priority="10"/> </failoverdomain> </failoverdomains> <resources> <ip address="XX.XX.XX.174" monitor_link="1"/> <nfsexport name="NFShome"/> <fs device="/dev/mapper/NFS_homes" fsid="2" force_fsck="1" force_unmount="1" fstype="ext3" mountpoint="/mnt/nfshome" name="nfs_homes" self_fence="0"/> <nfsclient name="node01" options="rw" target="node01.domain"/> <nfsclient name="node02" options="rw" target="node02.domain"/> <nfsclient name="node03" options="rw" target="node03.domain"/> <nfsclient name="nfs" options="rw" target="nfs.domain"/> </resources> <service autostart="0" domain="failover_nfshome" exclusive="0" name="client_nfshome" recovery="restart"> <ip ref="XX.XX.XX.174"/> <fs ref="nfs_homes"> <nfsexport name="nfshome"> <nfsclient ref="node01"/> <nfsclient ref="node02"/> <nfsclient ref="node03"/> <nfsclient ref="nfs"/> </nfsexport> </fs> </service> </rm> [1] http://sources.redhat.com/cluster/doc/nfscookbook.pdf Best Regards -- Denis Braekhus ------------------------------ Message: 6 Date: Thu, 15 Jan 2009 13:48:42 +0100 From: "Alain.Moulle" <[email protected]> Subject: [Linux-cluster] cman-2.0.98-1.el5 / question about a problem when launching cman To: [email protected] Message-ID: <[email protected]> Content-Type: text/plain; charset="iso-8859-1" Hi , About this problem, I wonder if it is a definitive behavior considered as normal ? or if this will work differently in a next release of cman or openais ? (in previous versions with cman-2.0.73, we did not had this problem) Thanks if someone could give an answer... Regards, Alain > Release : cman-2.0.98-1.el5 > (but same problem with 2.0.95) > > I face a problem when launching cman on a two-node cluster : > > 1. Launching cman on node 1 : OK > 2. When launching cman on node 2, the log on node1 gives : > cman killed by node 2 because we rejoined the cluster without a full > restart -------------- next part -------------- An HTML attachment was scrubbed... URL: https://www.redhat.com/archives/linux-cluster/attachments/20090115/dd62c 325/attachment.html ------------------------------ Message: 7 Date: Thu, 15 Jan 2009 08:59:56 -0500 (EST) From: Bob Peterson <[email protected]> Subject: Re: [Linux-cluster] Documentation enhancement requests/patches? To: linux clustering <[email protected]> Message-ID: <40063497.1429461232027996695.javamail.r...@zmail02.collab.prod.int.phx2 .redhat.com> Content-Type: text/plain; charset=utf-8 ----- "denis" <[email protected]> wrote: | Hi, | | After getting to know the "Configuring and Managing a Red Hat | Cluster" | documentation [1] fairly well, I have a few enhancement suggestions. | What is the best way to submit these? | | [1] | http://www.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/5.2/html/Clust er_Administration/ | | Regards | -- | Denis Braekhus Hi Denis, Probably the best way to do this is to open a new bugzilla record against product Red Hat Enterprise Linux 5, component "Documentation--cluster" If you have permission to look at it, you can follow this example: https://bugzilla.redhat.com/show_bug.cgi?id=471364 You can assign it to [email protected] or [email protected]. Regards, Bob Peterson Red Hat GFS ------------------------------ Message: 8 Date: Thu, 15 Jan 2009 15:01:17 +0000 From: Chrissie Caulfield <[email protected]> Subject: Re: [Linux-cluster] cman-2.0.98-1.el5 / question about a problem when launching cman To: linux clustering <[email protected]> Message-ID: <[email protected]> Content-Type: text/plain; charset=ISO-8859-1 Alain.Moulle wrote: > Hi , > About this problem, I wonder if it is a definitive behavior considered > as normal ? > or if this will work differently in a next release of cman or openais ? > (in previous versions with cman-2.0.73, we did not had this problem) > Thanks if someone could give an answer... > Regards, > Alain >> Release : cman-2.0.98-1.el5 >> (but same problem with 2.0.95) >> >> I face a problem when launching cman on a two-node cluster : >> >> 1. Launching cman on node 1 : OK >> 2. When launching cman on node 2, the log on node1 gives : >> cman killed by node 2 because we rejoined the cluster without a full >> restart > Alain, I'm sure this question has been answer many times on IRC and on the mailing list, as well as in the FAQ. Chrissie ------------------------------ Message: 9 Date: Thu, 15 Jan 2009 15:05:33 +0000 From: Chrissie Caulfield <[email protected]> Subject: [Linux-cluster] Re: [Openais] cman in RHEL 5 cluster suite and Openais To: [email protected] Cc: linux clustering <[email protected]> Message-ID: <[email protected]> Content-Type: text/plain; charset=UTF-8 Vivek Purohit wrote: > Hi Steve, > Thanks for the previous reply. > > I was able to run the checkpointing tests in the tarball Openais > on RHEL 5. > > I explored and came to know that the CMAN service of RHEL 5's > clustersuite runs as aisexec; thus the tests could be run directly. > > Can you please tell how the Openais is being used by RHEL 5's > CMAN service. > Hi, You might like to read these two documents: http://people.redhat.com/ccaulfie/docs/aiscman.pdf http://people.redhat.com/ccaulfie/docs/CSNetworking.pdf -- Chrissie ------------------------------ Message: 10 Date: Thu, 15 Jan 2009 09:08:08 -0700 From: Gary Romo <[email protected]> Subject: [Linux-cluster] GFS/clvmd question To: [email protected] Message-ID: <of48ae2273.1eee35b0-on8725753f.0057ec20-8725753f.0058a...@us.ibm.com> Content-Type: text/plain; charset="us-ascii" Why can't I mount my gfs logical volume on the second node in the cluster? I am creating a new GFS file system on an existing cluster. Here is what I did; 1. I determined I had space in an existing volume group (both nodes) 2. I created my logical volume (node 1) 3. I ran my gfs_mkfs (node 1) 4. I mounted my new lv on node 1 only Here is the error I get on node 2 # mount /gfs/new_mount /sbin/mount.gfs: invalid device path "/dev/vggfs/new_lv" I see that the logical volume is "inactive" on node2 and "ACTIVE" on node 1 inactive '/dev/vgclgfs/new_lv' [25.00 GB] inherit ACTIVE '/dev/vgclgfs/new_lv' [25.00 GB] inherit What do I need to do in order to make this logical volume active on node 2 ? I thought that this would have happened automagically via clvmd, and not have to be done manually. Gary Romo -------------- next part -------------- An HTML attachment was scrubbed... URL: https://www.redhat.com/archives/linux-cluster/attachments/20090115/1be5b 3b7/attachment.html ------------------------------ -- Linux-cluster mailing list [email protected] https://www.redhat.com/mailman/listinfo/linux-cluster End of Linux-cluster Digest, Vol 57, Issue 14 ********************************************* -- Linux-cluster mailing list [email protected] https://www.redhat.com/mailman/listinfo/linux-cluster
