Questions tagged [cluster]
A cluster consists of a set of connected computers that work together so that they can be viewed as a single system.
188
questions
0
votes
1
answer
117
views
Is it essential to use clusterwares for active-active cluster using SAN?
I want to use SAN storage as shared disk for two active nodes running on Linux, and I want to use it as raw device (without file system).
At first I thought it would have no problem if the disk is ...
- 1
1
vote
1
answer
985
views
YARN 3 and Spark: allocate a GPU
I can't find the working Spark option to require executors with a GPU.
I'm trying to setup a HADOOP cluster in order to run Machine Learning algorithms on available GPUs via Spark.
So far I'm trying ...
1
vote
0
answers
122
views
Over 100% objects are degraded, how this is possible?
Situation
Running a really small ceph cluster:
with three nodes (each containing mon, osd and mds)
an admin VM (containing also a mgr-service)
a pool replicating three times
Just to play around ...
- 121
-2
votes
1
answer
5k
views
Can I combine the computing power of two laptops for everyday use? [closed]
I have two laptops by different makers, both running Windows 10. Neither of them is comfortable enough for MS Office use or normal Web browsing. One has a better display, a quicker but smaller SSD, ...
- 97
0
votes
1
answer
76
views
No output and error file after running job in TORQUE
I installed mpich-3.2.1, torque-5.1.0 and maui-3.3.1 on Debian 9.
I have one master and one node.
I use NFS file system for sharing, and ssh passwordless.
When I run a job with a user that this user ...
- 1
0
votes
2
answers
2k
views
Terminating a SLURM zombie job
I have encountered the following problem during a first hard shutdown of the department cluster
I am in charge of.
The system is running SLURM 17.11 and using MariaDB/SQL to store accounting data.
To ...
- 101
0
votes
0
answers
163
views
pipenv module install fails with lmod modules enabled
Recently I've been experimenting with cluster computing/administration. I've got a small RasPi cluster with shared storage. On the NFS storage, Python 3.6 is installed to a folder. I'm using LMOD to ...
- 113
1
vote
0
answers
55
views
PostgreSQL Native Replication | Master - Master
I am configuring a high availability environment with Debian 9.4 machine, in which I have a Heartbeat working and need to include native PostgreSQL 9.6 replication, but by default this method of ...
4
votes
0
answers
662
views
Configuring Heartbeat to run shell script
I set up a high availability system with Heartbeat, I wish that when there was a drop in the main cluster and the secondary cluster was activated, Heartbeat start in it and execute a shell script.
...
3
votes
1
answer
1k
views
Hyper V fail over cluster
I have two Hyper V host. One with Intel adapter and other with Broadcom. Is this configuration supported.?
While running validation test I am getting error message.
Node 'HOST1.poc.lab' is missing a ...
- 31
0
votes
0
answers
199
views
Run the CPUMiner-multi mining package with the Mosix clustering package
When I run the CPUMiner-multi package for 'xmr' digital coin mining in Ubuntu Linux under the Mosix-4.4.4 clustering package with two nodes
$ ./mosrun -e ./minerd -a cryptonight -p stratum+tcp://...
- 1
0
votes
1
answer
41
views
Can I make two node Hadoop 2 Cluster on two different VirtualBoxes on two different lap-tops on the same network?
So I have two lap-tops with VirtualBox both running Ubuntu 18.4 and I am trying to set up Hadoop 2 Cluster. One VB to be the master and one VB to be the slave is it possible and how? Because all the ...
- 41
0
votes
1
answer
1k
views
Lots of disk errors but no hardware alert. Is this hardware related or software?
Summary: Random disks on one datanode of an Hadoop cluster keeps getting read-only. Jobs fails but there is no hardware alert on server.
Hello,
I am administring an Hadoop cluster which runs on ...
0
votes
1
answer
173
views
Backup MSMQ instance in a cluster
I found the right command that allows to backup a MSMQ instance in a server : mqbkup.
It is working for a local instance but when the MSMQ is in a cluster, I'm not able to make this command works. ...
- 101
1
vote
0
answers
23
views
Block outgoing traffic of cluster user
Recently I received an account in a computer cluster. It doesn't have a firewall so any kind communication can happen. However, I would like to set up a firewall for any outgoing traffic except for ...
- 11
3
votes
3
answers
4k
views
One operating system across two computers?
Is it possible to cluster two computers together and run one operating system across them? I have heard of Beowulf & such, but can I cluster things like PCI devices (graphics acceleration, ...
- 33
3
votes
0
answers
241
views
Zookeeper / Exhibitor cluster nodes keep restarting
I have successfully deployed 3 Zookeeper / Exhibitor nodes in Docker containers and they form a cluster. I am starting them via
java \
-jar /opt/exhibitor/exhibitor.jar \
--port ${EXHIBITOR_PORT} --...
- 141
0
votes
0
answers
96
views
Installing SSL on a Network Cluster
I am trying to install an SSL certificate on a distributed Windows cluster consisting of a domain controller and several web servers. Normally, on a set up where there's just a webserver, this has ...
0
votes
1
answer
52
views
troubleshoot pcsd ping resource on RHEL
we have configured ping resource monitor using REDHAT pcsd daemon.
In general - we get to error in /var/log/messages pointing to ping issue for specific entity (NAS server IP in this case). However ...
1
vote
1
answer
983
views
SLURM configuration: cons_res with CR_Core either cannot allocate resource or jobs end up in CG status
I am new to SLURM. I am trying to configure slurm in a new cluster.
I have 4 nodes each has 14 cores. I wanted to share nodes in a way that every core can run independently (i.e., node01 can have 14 ...
- 11
0
votes
1
answer
46
views
Different CPUs with same rating perform vastly differently... what is going on?
I built a cluster out of 6 OrangePi Zero's, which each have an H2+ Allwinner 1.2GHz quadcore CPU, 256MB RAM, and a 16GB class 10 Patriot SD card. They are connected via 100Mb ethernet switch and ...
- 1,219
0
votes
1
answer
58
views
Pipeline-init can not create directory
First of all, the pipeline-init command has the same function as the postgres initdb. The difference is that pipeline is a postgres fork.
dbastos@localhost ~/traderbot> pipeline-init -D /pipeline
...
5
votes
1
answer
2k
views
remove slurm sacct command double entries: "extern"
Jobs currently running show two entries, one of them has an .extern suffix. Completed (or failed) jobs also have a third entry: .batch. Is there a way to remove (or not show these) from the sacct ...
- 559
1
vote
1
answer
703
views
Creating users across all cluster nodes
Is there a way to add a user across all cluster nodes via command line? I have the manual instructions from IBM here (and listed below) and found instructions for listing the users (cl_lsuser), but ...
- 115
3
votes
1
answer
193
views
Create a Cluster Shared Volume with 3 Servers
I'm new to Failover Cluster and everthing that has to do with it. At the Moment I have two Hyper-V Hosts Connected to a VLAN and a third Server which I would like to use as a Fileshare so the VMs will ...
0
votes
1
answer
100
views
Simple Homebased "Super-kind-of" Computer [duplicate]
Well something similar to that.
Got an ancient desktop running a ubuntu-based distro and an equally ancient macbook.
Would like to connect and utilize both their computing capability for day-to-day ...
- 341
0
votes
2
answers
516
views
How to make MySQL cluster completely failsafe?
So I want to set up a MySQL database server with high availability - thus I got myself two devices (Banana Pi) and followed a tutorial on how to install the MySQL cluster software. I know that a ...
- 1
1
vote
2
answers
28k
views
Ping only working in one direction
I have a small cluster running centos 7 composed of machines A, B and C. They are all separate physical machines connected on the same LAN.
Machine A can ping B and C.
Machine B can ping A and C.
...
- 11
0
votes
1
answer
209
views
What does the output of `showq -r -u <username>` mean?
On a cluster, I did
showq -r -u matthey
where matthey is my username. Here is the output
active jobs------------------------
JOBID S PAR EFFIC XFACTOR Q USERNAME ACCNT ...
- 2,641
0
votes
1
answer
33
views
Hardware failures while building new cluster
Someone at my company is building a high-performance cluster (50 CPU cores, half a dozen machines, 32 memory modules per machine). We aren't experienced with clusters at all, and we are concerned that ...
- 11
0
votes
1
answer
3k
views
IIS failover Clusters?
I'm new to servers and stuff(I'm a programmer not a server guy) and I have a question about IIS failover clustering. I have two IIS servers up and running, and they're identical. I want to set up the ...
0
votes
1
answer
1k
views
Exchange Server DAG and Failover Cluster with High Availability
I have 2 exchange servers (EXCH1 and EXCH2) that have Exchange 2013 on them. They are both on Windows Server 2012 R2 (Standard) machines. They are also a part of a DAG (Database Availability Group) (...
- 115
1
vote
0
answers
52
views
Simple centralized logging for small Debian-based cluster
I have small cluster with 8 servers. Every one has Debian 8.4 installed. I'm only one administrator. What would be the simplest solution for centralized logging with following requirements:
Easy to ...
3
votes
1
answer
6k
views
Failover clustering validation report error: Validate Active Directory Configuration
I am deploying failover clustering on windows server 2012 r2 enterprise
When validate configuration, failover cluster validation report an error
At system configuration > Validate Active Directory ...
- 41
1
vote
1
answer
383
views
`qsub` leaves "NotQueued " in "blocked jobs"
I am working on a remote cluster (westgrid > bugaboo) on which I submit processes via a .pbs file. My .pbs looks like this
#!/bin/bash
#PBS -l procs=1
#PBS -l walltime=100:00:00
#PBS -N ...
- 2,641
0
votes
1
answer
403
views
Silent error when Terraform deploys to ECS
I’m new to Terraform and having problems setting up a very basic configuration. I want something that brings up my “docker-whale” image. From looking at Hashicorp and AWS docs, it seems I need an “...
- 151
1
vote
0
answers
292
views
Corosync dependency on unmanaged service
I'm trying to build an HA syslog cluster.
UDP syslog being stateless, there is no way for a balancer to ensure liveness of the nodes, so I'm experimenting with PCS/Corosync on CentoOS7.
Ideally, I ...
- 211
0
votes
1
answer
2k
views
Clustered SQL database requires full recovery mode?
The application we provide customers with requires an SQL database. Due to the nature of the application, we recommend the recovery mode to be set to Simple as it is not critical data that is stored. ...
- 595
0
votes
1
answer
2k
views
How to combine 2 Linux PC to serve Owncloud server
I just create owncloud server using my old PC with Intel Quadcore, 8GB memory, Centos 7, Apache, Mysql database. I already install RAID drive to make the read speed more faster. However, I try to ...
- 1
1
vote
0
answers
871
views
DiskView reports unallocated cluster, but WinHex reports cluster occupied by disk file - why?
I've recently been tinkering with deleting text files and seeing what information I can gather from the clusters that those files used to be on. However, I've hit something that's really made me ...
- 11
1
vote
1
answer
40
views
How to know how much (and which) data I have lost after disk failure
One of the disks in my cluster (with a RAID 5) died last week and I had to replace it. After a rebuild and a consistency check, some data belonging to different users were lost. I need to know exactly ...
- 123
1
vote
0
answers
156
views
Red Hat High Availabilty Oracle Resource Wallet Error
Environment: Running Red Hat 7.2, using Pacemaker 1.1.13-10.el7_2.2 and Corosync 2.3.4-7.el7_2.1 to implement cluster failover of an Oracle 12c Enterprise Edition Release 12.1.0.2.0 Database.
...
- 143
-3
votes
3
answers
5k
views
Can i setup cluster for windows 7?
I have two computers, both crummy ones, one is Celeron with two cores 2.2. second one is Pentium 4 with one core, both with 2 GB of ram. now what i want to achieve is that to use that second one to ...
user565801
1
vote
1
answer
3k
views
How to undo vgchange --clustered y?
At the beginning of a cluster setup I invoked sudo vgchange --clustered y [name] and accepted the warning that the volume might be inaccessible - no problem, of course, because the cluster isn't setup ...
- 2,272
-2
votes
1
answer
558
views
Is extra CPU on USB dongle possible? [closed]
Is it theoretically possible to build a local cluster for distributed computing using any type of a dongle plugged into a USB or HDMI port, and connecting to the Linux console inside over SSH just ...
2
votes
2
answers
174
views
two screen layer: reconection via ssh
my question is how to best manage two layers of "screen" sessions:
In my institute we have a compute cluster for numerical simulation. To login on that cluster one has to go via a load balancer who ...
- 23
5
votes
0
answers
689
views
Diagnosing iSCSI bottleneck
I just got my hands on a few DL380 G5s, and I thought I would use them in my home lab to test out the creation of a Hyper-V cluster using an iSCSI storage setup. I have installed Server 2012 R2 on all ...
- 427
2
votes
1
answer
184
views
Extensible Linux Cluster Operating System [closed]
Assume I have a laptop(2 cores, 4G memory, 512G regular disk) running ubuntu and a desktop(4 cores, 16G memory, 512G ssd disk) with no operating system installed(i.e. a new suite of computing ...
- 121
0
votes
0
answers
46
views
Accessing data on a cluster through ssh
I have compute access on a mac pro, and am planning to get storage access on a linux based cluster to which I can ssh into.
Is it possible to use the mac pro for my compute needs, but keeping the ...
- 53
4
votes
2
answers
6k
views
Upgrading gcc for a specific user account
I work on a computing cluster running CentOS6/RHEL6 that I share with ~100 other users, overseen by a single IT admin.
Recently, I started writing in C++, and some of my code depends on C++11. ...
- 143