Sponsored Content
Top Forums UNIX for Advanced & Expert Users Clustering solution for RH Linux AS and Solaris x86/AMD 64 Post 64606 by TioTony on Tuesday 1st of March 2005 11:54:53 PM
Old 03-02-2005
Hi Izzy,
I have various levels of experience with Sun Cluster, Veritas Cluster, HACMP (IBM), Red Hat AS clusters, Tru64 clusters, and SuSe clusters.
My personal opinion (please don't turn this into an OS vs. OS string) is Sun is dying and linux is rising. That being said it may still not be time for you to jump ship yet.
Wait until RH releases the Cluster Suite and GFS for RHAS 4.0 then get a copy to play with. RHAS 4.0 includes LVM and has an LVM gui which may make it easier to transition between Veritas terminalogy and Red Hat LVM. Red Hat's LVM is very similar to HP's LVM. RHAS 3.0 has LVM but it only gives you a GUI during the install. This may or may not be a problem for you. After the install it's all command line.
I have a few clusters running OCFS and GFS. My preference from the sysadmin side is for GFS because it too contains some LVM type features that allow you to grow the size of pools dynamically. Both GFS and OCFS were tough to install and configure when they were first released but they have both gotten better. OCFS is actually pretty easy to install and configure at this point. GFS is a little more cumbersome because you have to define the fencing mechanisms, nodes, and pools. Once you figure it out it is not so bad.
The only show stopper type of gotcha I have run into with RHAS clusters, OCFS, and GFS is a bug in GFS where the first 8 characters of the hostname must be unique. This is not fixed as of GFS 6 Update 4 but RH has told me it will be fixed in the next major release (6.1?). The only way I got around this was to rename my machines.
With the Veritas cluster we had a storage compatiblity issue with our HP VA 7410. We switched the storage to a SUN T3 and got around that problem.
Other then those two issues both have been pretty stable. RH will end up being cheaper becuase you don't have to pay for Veritas. The hardware and OS costs probably won't change much from Sun.
We have multiple RHAS clusters from versions 2.1 to 4.0 with 2,3, and 4 nodes. We have several clusters running 9i RAC and 10g ASM on the same machine using ext3, RAW (and ASM managed RAW), OCFS, and GFS without a problem.

Bottom line - RHAS 3.0 was really starting to be a solid production level clustering system. RHAS 4.0 appear to be even better (but I am still waiting for cluster suite and GFS).

If anyone wants working samples of the config files for GFS just post back and I will open a new thread with them.

Thanks,
TioTony
 

9 More Discussions You Might Find Interesting

1. SuSE

Linux suse for AMD

Hi everybody I have a SLES 8 on my own desktop and now I want to install oracle on it so I should install service pack3 on it, but when I setup it my linux Os can't start up and it say "Kernel panic". My desktop's components is 1-CPU:AMD Athlon 3200+ 64bit 2-Mainboard :Asus A8N SLES... (2 Replies)
Discussion started by: hkoolivand
2 Replies

2. Solaris

Solaris 9.0 (x86 on AMD Opteron) reboots after i FTP through Telnet

Hi All, I have installed Solaris 9 on on an AMD Opteron based server. The server is rebooting after i perform the following: - Logon to Solaris 9 AMD box through telnet or putty - ftp to any other box suppose box A where the FTP server is running - After logging in to the box A through... (1 Reply)
Discussion started by: jsvarma
1 Replies

3. Solaris

Solaris on AMD 770 chipset

hi is there anyone out there who has experience with installing Solaris on this chipset? there is one entry in the HCL doc about it (BigAdmin - HCL: Gigabyte GA-MA770-DS3 rev1.0 ) but that person apparently doesnt reply when another asks him some questions. im looking to install on... (3 Replies)
Discussion started by: Landser
3 Replies

4. Red Hat

Clustering Red Hat Enterprise linux Advanced Platform 5.2 (x86-64 bit)

Hello Professionals, We have high-end HP rack servers. We need to provide application fail-over for business continuity. We have done benchmarking of the application on RHEL 5.2 on HP servers to estimate the hardware requirement so as to meet next 3-5 years business growth. Presently we... (1 Reply)
Discussion started by: kgayyar
1 Replies

5. Solaris

Sample solution for Firefox-3.6.10 pkg in solaris 10 for X86

download pkg from sunfreeware web site (root user mode) step1: #mkdir opt/sft step2: download the package and save it in the opt/sft directory step3: #bunzip2 firefox-3.6.10.en-US.solaris-10-fcs-i386-pkg.bz2 step 4 #pkgadd -d ./firefox-3.6.10.en-US.solaris-10-fcs-i386-pkg ... (3 Replies)
Discussion started by: alamin2010
3 Replies

6. UNIX for Dummies Questions & Answers

Linux Clustering

hi guys Some time ago I used Linux HA(Heartbeat) to setup like 3 cluster. Now I have to install another 2 cluster and was checking more info to be sure HA was still used but I found some other stuff like OpenAIS - Corosync - Pacemaker to tell you the truth I am kinda confused here I get... (0 Replies)
Discussion started by: karlochacon
0 Replies

7. Linux

Linux os clustering

Hi, I have done the OS clustering in linux redhat 5.6, my one node is down and when i am trying to reboot the other node it is not coming up. any pointer to this would be helpful. the SAN storage luns are not coming as mounted (2 Replies)
Discussion started by: mohitj.engg
2 Replies

8. Ubuntu

Newbie With Linux Clustering, Need Help!!

Hi All, I am new user here and a new one to try clustering with Ubuntu nodes, and need help. If I should be in another place please mention. I have a two nodes with Ubuntu 14.04 installed on them. I need to make a cluster consisting of these two nodes with purpose of experimentation with... (3 Replies)
Discussion started by: IncognitoExpert
3 Replies

9. Solaris

Is it safe to install x86 Solaris 10 U6 after installed-Linux-and-FreeBSD?

I've installed Slack 14.2 on /dev/sda1 (/dev/sda2 is swap) and FreeBSD 12 on /dev/sda3 and lilo is the boot manager. FreeBSD slices are as follows; / on /dev/ada0S3a, swap on /dev/ada0s3e, /var on /dev/ada0s3b, /tmp on /dev/ada0s3d and /usr on /dev/ada0s3f. I hesitate to install Solaris 10... (2 Replies)
Discussion started by: vectrum
2 Replies
fsck.gfs2(8)						      System Manager's Manual						      fsck.gfs2(8)

NAME
fsck.gfs2 - Offline GFS and GFS2 file system checker SYNOPSIS
fsck.gfs2 [OPTION]... DEVICE WARNING
All computers must have the filesystem unmounted before running fsck.gfs2. Failure to unmount from all nodes in a cluster will likely result in filesystem corruption. DESCRIPTION
fsck.gfs2 will check that the GFS or GFS2 file system on a device is structurally valid. It should not be run on a mounted file system. If file system corruption is detected, it will attempt to repair the file system. There is a limit to what fsck.gfs2 can do. If important file system structures are destroyed, such that the checker cannot determine what the repairs should be, reparations could fail. GFS2 is a journaled file system, and as such should be able to repair damage to the file system on its own. However, faulty hardware has the ability to write incomplete blocks to a file system thereby causing corruption that GFS2 cannot fix. The first step to ensuring a healthy file system is the selection of reliable hardware (i.e. storage systems that will write complete blocks - even in the event of power failure). Note: Most file system checkers will not check the file system if it is "clean" (i.e. unmounted since the last use). The fsck.gfs program behaves differently because the storage may be shared among several nodes in a cluster, and therefore problems may have been introduced on a different computer. Therefore, fsck.gfs2 will always check the file system unless the -p (preen) option is used, in which case it fol- lows special rules (see below). OPTIONS
-a Same as the -p (preen) option. -f Force checking even if the file system seems clean. -h Help. This prints out the proper command line usage syntax. -q Quiet. -n No to all questions. By specifying this option, fsck.gfs2 will only show the changes that would be made, but not make any changes to the filesystem. -p Preen (same as -a: automatically repair the file system if it is dirty, and safe to do so, otherwise exit.) Note: If the file system has locking protocol lock_nolock, the file system is considered a non-shared storage device and the fsck is deemed safe. However, fsck.gfs2 does not know whether it was called automatically from the init process, due to options in the /etc/fstab file. Therefore, if the locking protocol is lock_dlm and -a or -p was specified, fsck.gfs2 cannot determine whether the disk is mounted by other nodes in the cluster. Therefore, the fsck is deemed to be unsafe and a warning is given if any damage or dirty journals are found. In that case, the file system should be unmounted from all nodes in the cluster and fsck.gfs2 should be run manually without the -a or -p options. -V Version. Print out the program version information. -v Verbose operation. Print more information while running. -y Yes to all questions. By specifying this option, fsck.gfs2 will not prompt before making changes. fsck.gfs2(8)
All times are GMT -4. The time now is 02:10 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy