A computer cluster consists of a set of loosely connected computers that work together so that in many respects they can be viewed as a single system.

learn more… | top users | synonyms (1)

1
vote
0answers
5 views

Is there a way to check resource utilization on a cluster running SLURM?

Not sure if this question fits better here or at Unix.SE but I figured it feels more appropriate here. I am running some computation-heavy research on a national cluster which uses SLURM for ...
0
votes
0answers
21 views

ElasticSearch Cluster fails 5 minutes after starting on Azure

I am trying to build an ElasticSearch cluster on Azure. I have done it successfully for testing purposes with 3 VM's under the same virtual network. It functioned very good. Because of my ...
0
votes
0answers
19 views

Private IP Addresses not shown for blades

I have 2 IBM bladecenters with a total of 9 blades. I am currently setting them up to do cluster simulations. I installed ubuntu 14.04 on all of them. Also, the blades are not connected to the ...
1
vote
2answers
53 views

Replacing failed hardware in a hyper-converged cluster node [closed]

Is it possible to replace failed physical components (such as disk, RAM, or power supply) in a hyper-converged node (such as VXrail or Nutanix) without taking the node offline?
0
votes
0answers
27 views

Can't ssh from nodes to VM Proxmox 4.2

SOLVED: I edited the /etc/ssh/sshd_config and changed "PermitRootLogin" to "yes" instead of "without-password". For whatever reason by default it is set to yes for each node, but in the VM it's set ...
0
votes
1answer
14 views

How does spinning up a node within a cluster reduce CPU load?

I have a VM (representing a baremetal server) with a load balancer and a cluster of 3 container nodes that requests distribute to. Say some threshold is hit and I scale up a 4th node in the cluster. ...
0
votes
0answers
31 views

“Authetication key already exists” error while adding computer to cluster proxmox

I had 2 computers running proxmox which were set up in a cluster called "cluster1". For simplicity the computers were named computer1 and computer2. computer1 was the primary, and computer2 was added ...
0
votes
3answers
107 views

How to SSH to all instances in EC2 cluster?

I just created a cluster using spark-ec2. I followed steps here http://spark.apache.org/docs/latest/ec2-scripts.html . Now I want to configure all slaves (150), i.e., install Java 8 for both master ...
-1
votes
1answer
19 views

MongoDB, How can I migrate database to new cluster with higher version without an outage

We are having our mongodb cluster at AWS and our production DB mongo is very old - we are still using version 2.6 then latest version at the moment is 3.2.7. We need to upgrade mongodb binaries to the ...
0
votes
2answers
87 views

Unable to communicate with pacemaker host while authorising

I'm trying to configure pacemaker cluster with two hosts, I'm using two centos 7(CentOS Linux release 7.2.1511 (Core)) virtual machines. What I did so far: I installed packages: yum install ...
0
votes
0answers
22 views

How can I launch graph500 benchmark on nVidia GPU with CUDA?

How can I launch graph500 benchmark on nVidia GPU with CUDA? There is written: https://devblogs.nvidia.com/parallelforall/gpus-graph-predictive-analytics/ At the IEEE Bigdata conference later in ...
0
votes
1answer
18 views

Kubernetes cluster ip not answering

We've setup a kubernetes cluster with 3 masters and 3 workernodes. Then we've installed the kubernetes-dashboard which failes because it can't connect to kubernetes (api-server). It's looking for ...
1
vote
1answer
77 views

Postgresql replication with pgpool or pgbouncer

I'm looking for a good configuration for a postgresql replication with a robust failover strategy (self hosted). Actually, I configured two postgresql instances in master / slave with repmgr. Now, I ...
0
votes
0answers
34 views

Rancher usage of mariadb galera cluster

I was deploying mariadb galera cluster at ranhcer environment and have a problem of node hosts: 6/3/2016 7:11:39 PM160603 15:11:39 mysqld_safe WSREP: Recovered position 00000000-0000-0000-0000-...
3
votes
3answers
90 views

HyperV Failover Cluster (with/without CSV)

Been browsing a few websites, and it seems right now all hyper-v clusters are build on CSV ( clustered share volume ). I have a current setup (2 node window failover cluster). I have 2 vms running ...
2
votes
0answers
32 views

LSF nodes die after exhausting all memory

I have a LSF cluster and experience the following problems. After some time, random OS processes start to die (they become zombies or defunct), and the LSF nodes do not report any troubles back to ...
0
votes
1answer
31 views

Slurm: Have two separate queues for GPU and CPU-only jobs

At the moment, we have set up Slurm to manage a small cluster of six nodes with four GPUs each. That has been working great so far, but now we want to utilize the Intel Core i7-5820K CPUs for jobs ...
0
votes
1answer
64 views

Four node asterisk cluster

I have a question regarding Asterisk servers. I have a client that have two locations connected with fiber optic. On both locations they have PRI ISDN-s. On one location there are 4 PRI ISDN, and on ...
0
votes
1answer
44 views

Unison / file syncing between servers without using root account

I have setup 3 servers one with a load balancer and the other 2 will host the websites files and I have a replicated MySQL database on both too. I have installed ispconfig on both and it's set to ...
0
votes
2answers
71 views

How can I run Spark on a cluster using Slurm?

I have written a program example.jar which uses a spark context. How can I run this on a cluster which uses Slurm? This is related to http://stackoverflow.com/questions/29308202/running-spark-on-top-...
0
votes
0answers
40 views

How can I configure a single-node cluster on RHEL 7?

For particular purposes, I'd like to know how can I configure a RHEL 7 HA cluster with only one node (of course, no HA will be available in this scenario). The following article says that single-node ...
0
votes
0answers
29 views

How to allocate drives with SLURM

I need to share drives (SSDs or hard drives) in a compute node. The Generic Resource (GRES) Scheduling is a Slurm module designed to manage different kind of resources. For instance it gets a plugin ...
1
vote
0answers
38 views

Percona XtraDB Cluster 5.6 does not start

All the good days. I want to run for test purposes Percona XtraDb Cluster on ubuntu 14.04. The basis here took these two articles https://habrahabr.ru/post/152969/ https://www.percona.com/doc/...
0
votes
1answer
31 views

Error AMQ4036 when trying to add a QueueManager to a Cluster in IBM MQ?

I am trying to setup a MQ Cluster that has the following config There are 3 host machines, ubuntuvm-1, ubuntuvm-2 and ubuntuvm-3 ubuntuvm-1 has one Full repository queue manager and one partial ...
2
votes
0answers
39 views

Lustre alternative for 70-node scientific cluster

Our lab has a cluster with 70 compute nodes 4 IO nodes Infiniband QDR interconnection 12T disk array accessed using IB SRP protocol The major application is debugging and running MPI-based ...
1
vote
1answer
77 views

Is there a difference between LVM and HA-LVM or is it just how you use it.

I am studying up on HA Clusters using pacemaker and corosync. The Red Hat documentation treats LVM, HA-LVM and CLVM as separate topics. I understand the added features and benefits of CLVM over HA-...
0
votes
1answer
23 views

Mongodb sharded cluster architecture

I have a mongodb cluster on aws. In order to reduce network latency I want to have multiple shards in different regions. In each region there will be the application nodes that connect to the ...
0
votes
0answers
24 views

mysql ndb cluster - Management node and data node go down together

I have an NDB scenario with the following details: Blade 1: VM 1 : Management Node (Active) VM 2 : Data Node VM 3 : SQL Node Blade 2: VM 1 : Management Node VM 2 : Data Node VM 3: SQL Node If ...
0
votes
1answer
64 views

Windows Server 2012 WSFC and NLB together

I am looking to run a combination of WSFC and NLB in my service stack. I have a stateless service that I want to load-balance with NLB, and several "singleton" services that need to be run with fail-...
0
votes
2answers
29 views

Is it necessary to secure TCP traffic for PHP FPM over shared private VPS network?

There are lots of recommended practices and tutorials for securing various components of a LEMP/LAMP stack when running in a cluster whose private network traffic is shared by other, potentially ...
0
votes
1answer
61 views

Passing SAN Storage to Virtual Machine in active/passive HA Cluster

I need to attach a fibre channel SAN disk to one virtual machine that resides on an active/passive cluster (pacemaker, corosync). As it is the backup storage, corruption has to be strongly avoided. ...
4
votes
0answers
65 views

RabbitMQ two node high availability cluster intermittently breaks with Veeam backup jobs

The environment consists of two 2012R2 Virtual Machines running RabbitMQ in high availability (ha-all) on their Queues. I use Veeam to create snapshot backups that are sent offsite as part of the DR ...
0
votes
0answers
19 views

PowerHA 7 for AIX 6.1: Modifying the IP source for outgoing packages

I've clusterized a server which has ip address 10.10.10.10/24 (example ip) For that I added two new nodes with ips 10.10.10.11 and 10.10.10.12 and changed the ip 10.10.10.10 to the service IP address....
1
vote
1answer
94 views

ceph - can't start osd on rebooted cluster host

I've rebooted the server (one of ceph's hosts). Started the cluster, but the osd, that's on the host, which was rebooted is down. The osd's # is 2, so when I try: sudo /etc/init.d/ceph start osd.2 ...
1
vote
0answers
14 views

Opendaylight Cluster : Error resolving artifactorg.jolokia:jolokia-osgi:jar:1.1.5:Could not transfer artifact

I'm using the integration distribution of Karaf (0.4.1-Beryllium) to deploy a 3 node cluster with Opendaylight Controller. After using the deploy.py script and checking all the configuration files on ...
0
votes
0answers
44 views

windows 2012 r2 guest OS cluster in SUSE KVM environment steps

We are trying to configure the windows 2012 R2 guest OS cluster in SUSE KVM environment we are getting message no disk were found on which server we need to perform cluster validation test. As we ...
0
votes
1answer
80 views

Can't start VM OpenNebula

I have 3 nodes with CephRBD + OpenNebula + Corosync + Pacemaker + Galera + crmsh. Software: OpenNebula 4.14.2, CentOS 7.1 All works, but when I instantiate VM, a get an error and VM in PENDING ...
1
vote
4answers
70 views

Good generic term for a 'domain' or 'network boundary'

I'm trying to find a term...and its harder to find than I thought. Given a group of computers are within a network boundary, and have a set of user accounts (system admins) setup to oversee computers ...
-1
votes
1answer
24 views

Servers, cloud and network

I need so help so I have few questions. I work in small company and we are going to be buying one or two servers for our needs. We are in area that isn't well equip with internet (only ADSL or WiMAX) ...
0
votes
0answers
20 views

Rabbitmq clustering - Client configuration

Let's say I have setup a rabbitmq cluster on node1 and node2. Node1 is master and other is slave. Now how should we configure the clients? Do we need a load balanced hostname which will distribute ...
0
votes
2answers
81 views

window server 2012 r2 clustering - shared storage required or not

I just came to know that we can setup a 2 node window failover cluster without using shared storage (SAN) In that case, how does services running on 1 node failover to another when their data is not ...
3
votes
2answers
103 views

Hyper-v failover or replication in 2 node cluster

I'm trying to setup some form of a failover cluster between 2 hosts. I want to be able to have high availability on 2 of my VMs. It seems that some form of shared storage is required for this to work ...
2
votes
1answer
40 views

In a Pacemaker cluster, is there any way to set start-failure-is-fatal on a per resource basis?

I've got it set cluster-wide currently, but I'd really like it to be set to 'true' by default and false for only my Slave DRBD resource. Using Pacemaker 1.1.12 and Corosync 1.4.8.
1
vote
3answers
101 views

Must all nodes of an Elasticsearch cluster be identical?

In an Elasticsearch cluster, must all nodes be identical i.e. having the same: Operating System Amount of RAM CPU speed and number of cores etc. or is it allowed for the nodes to be different? (...
2
votes
2answers
43 views

Proper way to distribute binaries to linux cluster

I recently had a disagreement with my boss over the proper way to distribute executable binaries across compute nodes in a Linux cluster. All the compute nodes are running the distro and the same ...
1
vote
1answer
38 views

Why won't 'docker run' on a swarm show output from a command?

I've successfully setup a Docker Swarm cluster consisting of one master and two nodes. Everything works as expected, only one thing deviates from the documentation. According to the docs (look for "...
1
vote
0answers
23 views

Performance concern with hosting large clustered DBs on VPS cloud servers

I wonder if anybody else bumped into this issue? You get a couple of beefy 12-core VPS instances in some cloud like DigitalOcean and set up a database on them in "active-passive" cluster ...
0
votes
1answer
44 views

CIB not supported: validator 'pacemaker-2.0', release '3.0.9'

I am trying to make two node clusters with pacemaker and corosync. When I want to change config with crm I get the following error: CIB not supported: validator 'pacemaker-2.0', release '3.0.9' ...
0
votes
1answer
132 views

Postgres server not starting on drbd + pacemaker

I am trying to bring up the postgres server over ubuntu 14.04 with drbd and fiesystem. Service state as below: Last updated: Mon Mar 14 01:16:45 2016 Last change: Mon Mar 14 01:05:53 2016 via ...
0
votes
0answers
10 views

Is it possible to run CMAN without configfs?

I run a custom 3.18.x kernel that does not have configfs enabled. Is it possible to run CMAN cluster management without configfs? If necessary it would be much easier for me to rebuild CMAN than it ...