discusion related to cluster mecanism.

learn more… | top users | synonyms

1
vote
2answers
428 views

Trying to install a cluster, DHCP failing to start

Trying to install a linux cluster using this tutorial. I am running CentOS 6.3. When I try to start DHCP with service dhcpd start, it fails. I've used tail -f /var/log/messages and dhcpd -t, but I ...
3
votes
1answer
66 views

SLURM: Custom standard output name

When running a SLURM job using sbatch, slurm produces a standard output file which looks like slurm-102432.out (slurm-jobid.out). I would like to customise this to (yyyymmddhhmmss-jobid-jobname.txt). ...
2
votes
1answer
125 views

Running GNU parallel --sem hangs on SGE Cluster

I'm trying to run 3 commands in parallel on an SGE cluster, e.g. cmds=("prog 1 2 3" "prog 4 5 6" "prog 7 8 9") for cmd in "${cmds[@]}" do sem -j 3 $cmd done sem --wait So, my commands execute ...
2
votes
1answer
265 views

Clustered NFS server reply ERR 24: Auth Bogus Credentials (seal broken)

I have 4 servers on the VirtualBox. Two of the servers are a CentOS 7 cluster with Pacemaker(corosync), and they have an NFSv4 server in Active/Passive mode. There are also 2 clients with CentOS 6, ...
2
votes
1answer
132 views

reduce bandwidth consumption from downloading yum updates

I have several machines (7) running fedora and would like to reduce the amount of external bandwidth I am consuming. I see that there were several attempts to setup mirrors, but those projects seem ...
2
votes
1answer
309 views

mpd daemon prematurely ending jobs

I am a little out of my depth here so bear with me. I am trying to configure mpirun and mpiexec to run software called Materials Studio on a 1 node, 2 processor, 12 core cluster. The submission scheme ...
1
vote
1answer
222 views

Failover NFS service with shared storage

I need to configure a cluster with shared storage that can be moved from Node A to Node B and vica versa. In case of failure of Node A, Node B should take over the IP address associated with the NFS ...
1
vote
1answer
129 views

Python processes appearing and disappearing in top

I concurrently launched around 12 Python scripts on a 24-core server (Ubuntu 14.04 LTS x64). Each of Python script takes around one hour to complete. I see in top that they appear quite sporadically, ...
1
vote
1answer
98 views

Cluster Lock Manager without TCP/IP

I have 2 linux sharing same SAN disk LUN. There are many cluster files systems to manage lock/unlock. But all of them are using a seperate TCP/IP comminication. But I need a cluster lock manager ...
0
votes
1answer
31 views

Terminal to manage cluster, where the nodes are accessible from a remote PC?

I need to manage multiple cluster nodes from my PC. The catch is that the cluster nodes are accessible only from a remote PC. So first I have to ssh to the remote PC, and then ssh to a cluster node ...
0
votes
1answer
32 views

How to see more information for queued Jobs in SLURM

With srun --test-only SLURM will return expected start date AND expected nodes. However, if the job is submitted, I cannot get the information about the nodes. Is there any way to find out, what ...
0
votes
1answer
579 views

Why did Red Hat drop support for DRBD beginning with RHEL 7?

Does anybody know the reasons why Red Hat decided to drop support for DRBD beginning with RHEL 7, thus favoring GlusterFS as the preferred solution for network-based replication / mirroring in a ...
0
votes
1answer
245 views

mysql high availablity for LAMP application (ubuntu 14.04)

im trying to achieve HIGH Availability (HA) for a LAMP application(News Portal website) which might be of interest to many unfamiliar users.For files im using Glusterfs (replica) for redundancy ...
0
votes
1answer
152 views

Powerful Workstation

I heard about ILM's (Industrial Light and Magic) workstations on which they create models and scenes for a lot of different movies. They have workstations with more than 1024 GB of ram and a lot of ...
0
votes
1answer
34 views

Is it possible to install RHEV on el6 clustered computer

Is it possible to use Linux (el6) clustered computer can be used as a hypervisor for RHEV 3.3/Ovirt 3.3? Cause, we can not combine RAM/CPU from two or more hypervisores to be considered as a one ...
0
votes
1answer
461 views

What software should I use to create a HPC (High performance compute) cluster with CentOS 6.5 VMs locally?

I've read about Torque, Slurm resource managers and Maui, Open Grid scheduler, Sonf of Grid Engine but I do not know which is more appropriate to use in a local environment? I've also read that I need ...
4
votes
0answers
63 views

Check if a remote machine is in a cluster

How can I verify if a remote machine is a cluster machine? I just created the following ssh command to verify if the remote machine is a VRTS cluster or a Linux availability Red Hat cluster ...
4
votes
0answers
199 views

Takeover delay in Heartbeat HA cluster

I have set up a heartbeat HA two node cluster. And both of server have same configuration of ha.cf #cat /etc/ha.cf logfile /var/log/ha-log autojoin none bcast eth1 warntime 20ms deadtime 40ms ...
3
votes
0answers
784 views

Corosync/Pacemaker cluster loss of synchronization

I'm facing a problem with the corosync and pacemaker cluster software. I've two virtual nodes (Scientific Linux 6.4) and corosync/pacemaker installed on them: corosync-1.4.1-17.el6.x86_64 pacemaker-...
2
votes
0answers
39 views

Running R scripts on a Linux cluster

I have access to an IBM IDataplex Cluster with CentOS v.6.2. If I want to run the following R script on R: library(data.table) library(mgcv) library(reshape2) library(dplyr) library(tidyr) library(...
2
votes
0answers
134 views

XFS over iSCSI share with cluster; will fencing cause data corruption/loss?

When using XFS on iSCSI share, where HA is enabled, nodes may be fenced by cluster. Will this result in data loss / corruption / etc. ?
2
votes
0answers
42 views

heartbeat stops after reviving primary node

I followed these steps correctly, everything worked well before I have revived heartbeat on node01 "at the last steps of the tutorial" when I hit /etc/init.d/heartbeat start it started correctly but ...
1
vote
0answers
13 views

How to disable automatic update of the correct GPU count for each MOM node in Torque?

I have small installation of Toque 4.2.9. It is compiled with --enable-nvidia-gpus option. According to the documentation, when this option is used nodes file if automatically updated with the correct ...
1
vote
0answers
13 views

qsub launch a job from node

I have a pipeline that is launched using a command. For some admin/server reasons, that tool has to be launched from a node. That tool triggers all the steps from my pipeline in a file called pipeline....
1
vote
0answers
34 views

Running web server without root access

Excuse the probably simple question, I don't know much about running web servers. I' am trying to run some software on a compute cluster I have access to on my school campus (so I don't have root ...
1
vote
0answers
54 views

FreeBSD 10.2 - Is three-way HAST replication possible?

I have a pair of FreeBSD 10.2 servers using HAST to keep a volume replicated between them. I'm introducing two new storage nodes that have much newer hardware and will eventually replace the existing ...
1
vote
0answers
694 views

qsub job in state Eqw error: can't chdir to directory: No such file or directory

I'm trying to submit a job on a cluster via qsub, but it gets stuck in state "Eqw" with error message: $ qstat -j 466 | grep error error reason 1: 09/18/2015 17:12:32 [1125:3453]: error: ...
1
vote
0answers
2k views

Error starting MySQL in a MariaDB Galera cluster in CentOS 7

I'm trying to setup a MariaDB Galera cluster following this tutorial exactly as it is, and I'm stuck in the Step 7– Initialize the first cluster node, I'm doing so with sudo /etc/init.d/mysql start --...
1
vote
0answers
697 views

clvm (cman) has problems with dlm

I am trying to create a volume group which is accessible on two servers with clvm. I read about it and it looks like it is most easily to do this with cman. But after hours of testing I cannot achieve ...
1
vote
0answers
1k views

About mem and vmem

I am working with a cluster machine running under linux. I have a shell script that uses mpirun to submit my jobs to the cluster machine. In that same script, I can choose the number of nodes that ...
1
vote
0answers
123 views

Microcomputer cluster for stress testing

I'm looking to build a microcomputer cluster at work for stress testing one of our applications, and while I have a little financial leeway, I would like to keep the cost low and the computing ...
0
votes
0answers
10 views

CentOS or RHEL 7 - change nic used by cluster

I have a two nodes RHEL 7.1 cluster (NFS server) that currently uses 10 Gbps ethernet nics. The servers use a SAS shared storage. If I have to switch to Infiniband (IPoIB), can I stop the cluster, ...
0
votes
0answers
61 views

Spark example Pi fails to run on yarn-client mode

I tried my spark app on yarn-client mode and it's failure. And then I tried to run spark example of Pi on it and failure too. The systems I'm working on are as following: Ubuntu 14.04.4 Java(...
0
votes
0answers
13 views

Improving MPI performances on a heterogeneous cluster

I am running some parallel programs on a heterogeneous cluster: 2 CentOS virtual nodes (node 1 & node 2) on a Mac host and 2 others (node 3 & node 4) on a PC. When I run a parallel program (...
0
votes
0answers
17 views

Any clustering / load balancing solutions for VPN servers that support external IPs?

I am looking for a clustering / high availability / load balancing solution (something like ucarp or clusterip, etc) where it allows the nodes to be external IPs. Specifically, I would like it for ...
0
votes
0answers
32 views

How to generate following output file by using PBS script language?

I need to run interacting job with such specifications by using PBS on the cluster. This job must write the output file, in which there is following information (I have not specify this in the .pbs ...
0
votes
0answers
40 views

How to bootstrap a new dependency configuration on a cluster?

I have access to a cluster running a CentOS 6, which have not been updated. For that reason, the glibc, ld-linux and other dependencies that, I think, are kernel-based are all out-of-date. I have no ...
0
votes
0answers
18 views

MySQL Cluster node crashed, how to re-add in cluster?

We have 2 data node in mysql cluster, and one of them got crashed (totally). I have re-install new OS and now i want to re-add same node back to cluster. How to do that without re-starting whole ...
0
votes
0answers
57 views

problem with ABAQUS job on Rocks cluster

I created ROCKS6 cluster and i installed ABAQUS on my cluster and it works well on frontend, but i can not submit job on my 2 other compute nodes. I just want to run ABAQUS on this cluster (frontend ...
0
votes
0answers
102 views

How to install ABAQUS software on Rocks cluster 6.2

First of all sorry for my english. I create a new cluster with 3 . which one of them is frontend and two of them is compute . i should notice that, i am newest in linux and linux hpc clustering. I ...
0
votes
0answers
13 views

Writing a unique number in a parallel ssh / clusterssh environment

I'm trying to set up a cluster of Kafka servers and I'm currently using clusterssh for it, which works great. As one step of the setup, I have to assign a unique id to every cluster (as in 0, 1, 2, ...
0
votes
0answers
18 views

High Performance Cluster setup

I am starting to setup a High Performance Cluster (HPC) that will have a head node, 16 sub nodes and 8 GPU nodes. I am looking into Red Hat 7.2 as they state to have HPC capability. Thoughts on ...
0
votes
0answers
225 views

How do I add a Redis Cluster to a Pacemaker CentOS 7 Cluster?

I need to set up a CentOS 7 web server cluster that is able to manage user sessions while also load balancing across the nodes in the cluster. Towards this end, I have transcribed instructions from ...
0
votes
0answers
18 views

spark job dependency on AMI version

I have run a spark job on emr cluster with AMI version 2.4.0-amzn-7 and it completes in 11 hours. but when same jobs are run on AMI 2.4.0-amzn-3, the process takes more than 18 hours to complete. ...
0
votes
0answers
134 views

HPL benchmark show very low calculation speed

I’m constructing cluster for mph parallel calculation. To check performance of my node, I used HPL linpack benchmark. However, the benchmark result is much more low than a theoretical value. The ...
0
votes
0answers
91 views

Getting error in nslookup

I am trying to do nslookup skhdphost1 but I am getting below error. When I try to Reverse lookup checking using command nslookup 172.16.xxx.xx again getting the same error: The error message is: [...
0
votes
0answers
364 views

rgmanager [fs] start_filesystem: Could not match “f6a58cf7-b39b-489c-add5-444ceab8af9d” with a real device

I am trying to install two node cluster for web server. We have super mico hardware HA BOX. It has 5 disk. one disk is used for node1 os installation and another disk is for node2 OS. Remaining disk ...
0
votes
0answers
372 views

HAProxy with Roundrobin working only on 2 of 3 MySQL servers, why?

Good day. I'm following this article to configure HAProxy to balance with roundrobin between 3 MySQL servers (db1, db2, db3) in a MariaDB Galera cluster (working perfectly). However I have a problem....
0
votes
0answers
26 views

centos cluster set core/each_program

I'm using unix server (OS: centos5) and I'm using several clusters. Some clusters have 8 cores, other are 24 cores and etc. I execute jobs using this command. qsub -q cluster_name -d $PWD ...