Questions tagged [cluster]

discussion related to cluster mechanisms.

Filter by
Sorted by
Tagged with
0
votes
0answers
6 views

Add new Storage Volume Resource to existing Redhat Cluster

We are using RedHat Pacemaker request OS 7.6 Pacemaker pacemaker-1.1.20-5.el7_7.1.x86_64 Now we have added New Lun to this cluster to keep the Oracle backups on it, This New lun will be part of ...
0
votes
0answers
26 views

How can I cancel all waiting jobs with qsub?

I am running a lot of jobs with qsub: some are running, some are waiting. Is there a way to cancel all the jobs for a given user which are queued/waiting without giving the individual job IDs?
0
votes
0answers
4 views

what is the difference between stonith and fence in linux 7 cluster

i'm just starting study linux cluster, noticed that stonith has action off and reboot, fence device also has action that can reboot/off device in cluster. so what's the diff? if they have diff action, ...
0
votes
0answers
22 views

how to merge resources of Multiple Servers?

i just need to know is this possible that we can merge or create the resource pool of multiple server , or how can we create a cluster of multiple server so that they act like a single Server with ...
2
votes
0answers
57 views

how to set passwordless authentication in a cluster where users /home directory from headnode is mounted to all machines /home of the cluster

First of all thank you in advance for your help. I hope the title makes sense. Basically, on the headnode the users' home directory (i.e: headnode:/home/eric) are NFS shared and mounted to all the ...
0
votes
1answer
141 views

Pacemaker apache resource is Failed to access httpd status page after change to HTTPS

I get this error from pacemaker after i change apache from http to https. now my ocf::heartbeat:apache resource is not find status page. I generate SSL certificate separately for 3 servers. Everything ...
2
votes
1answer
137 views

Simple cp command not working (inside shell script and submitted using sbatch on a cluster)

I am working on a cluster running RHEL and I submit jobs using the following command. sbatch MyScript.sh The content of the MyScript.sh are as below. #!/bin/sh # .... # Other SBATCH related commands ...
0
votes
0answers
8 views

What is the correct procedure for upgrading OS and application server on two-node failover HA Pacemaker cluster?

I've scoured enough search engine results before coming over. Looking for the correct procedure to upgrade an Ubuntu server and an application server on a two-node failover HA Pacemaker/Corosync/DRBD ...
0
votes
0answers
95 views

Mysql resource only starts when run by root user inside pacemaker cluster. Why?

I have a pacemaker/corosync cluster with a basic configuration, nothing relatively complex. (The mysql database is on a shared disk which is not important.) The mysql primitive: (ocf_heartbeat_mysql) ...
0
votes
0answers
31 views

iptables extension CLUSTERIP issue:

I have Debian Linux with Kernel 2.6.32 & IPTABLES v1.6, trying to test iptables CLUSTERIP module. Following steps were performed to build IPTABLES from the source: ./configure --disable-nftables -...
1
vote
1answer
33 views

compiling HPL-2.0_FERMIv15

when i compile xhpl i always get the error message: ./xhpl: error while loading shared libraries: libdgemm.so.1: cannot open shared object file: No such file or directory when i type ldd xhpl: linux-...
1
vote
0answers
116 views

Cannot seem to start pcs cluster (NFS Cluster) disk_fencing trouble

For the life of me, I can't find a clear answer on how to start my NFS active / passive cluster. I have two nodes, node1 and node2 and followed the guide here: https://www.linuxtechi.com/configure-nfs-...
0
votes
0answers
59 views

PCS cluster node going offline when adding 2 entries for a cluster node in /etc/hosts

I have updated the /etc/hosts file with 2 IP's for the host and PCS cluster is not in sync after reboot, # pcs status Cluster name: clusterha Stack: corosync Current DC: vm2 (version 1.1.19-8.el7_6....
0
votes
1answer
427 views

DRBD: “Couldn't mount device [/dev/drbd0] as /mydata” when failing over or rebooting node

I'm creating a cluster system using two ESXi hosts, with a CentOS 7 server on each. Going through I created the filesystem, and it mounts on node1. When I perform a standby or reboot from node01 to ...
0
votes
2answers
657 views

How to Remove caavg_private Properly on AIX?

I am trying to cleanup a server which had a PowerHA configuration. I have stopped cluster (smitty clstop) and removed resource groups. How do I remove the caavg_private properly? hdisk5 ...
0
votes
0answers
18 views

Virtual IP mapping on a physical standalone box

We have a clustered HP-UX server and few mounts are shared among these two servers and move along with fail-over. These two physical servers are redirected by a virtual IP. This virtual IP is used by ...
0
votes
1answer
16 views

RHEL 6.9 HA clustering issue with one node completely down

I have two node servers with SAN storage. Each node have RHEL 6.9 with HA and the partition are mapped from the storage using fiber cables with clustered resources. The thing is when the two nodes ...
0
votes
0answers
40 views

galera mysql single node recovery issue

We have 3 node galera cluster and it was functional until last night one of node crashed because of hardware failure and after fixing hardware i am not able to start mariadb and getting following ...
0
votes
0answers
196 views

How to identify the cause of reboot as the log was not captured in a SLES 12 SP3 (2 node cluster)?

Currently, a failover occured from node 1 to node 2 as there was a sudden reboot on node 1. Upon analyzing the log, it is found that the reboot log was not captured and also kdump was not configured ...
0
votes
1answer
73 views

Trying to Create an AFM relationship by using GPFS protocol. Having error in cache side cluster

I am trying to create an AFM relationship by using GPFS protocol. Having error in cache side cluster. Steps of Home cluster : 1) Create a home cluster (cluster name - gpfs01). 2) Create a file ...
1
vote
0answers
111 views

OpenLDAP Cluster

Trying to implement an OpenLDAP cluster, I already managed to set up the two backend LDAP servers in mirroring mode. The application (iRedMail) using the LDAP service is running on the same systems ...
1
vote
0answers
88 views

How does htop calculate times? (Impossible times shown)

I was running what was supposed to be a small make job on a small node in our cluster, but a sorting process seemed to be overwhelming the allocated RAM, so I killed it after 20 hours (same jobs on ...
0
votes
1answer
31 views

how to move fs resource in service group in ccs cluster

I have a ccs cluster running on RHEL 6.4 where there is no luci service, i have added a filesystem resource to the cluster with the below command but i need to move the resource inside a service group....
0
votes
1answer
59 views

How to setup a compute cluster in local network on top of linux

I have 3 machines in my local network running manjaro. I am running python scripts using dask, pandas, etc which max out the cpu on the first machine and I usually need to wait more than 30 min until ...
0
votes
2answers
81 views

Why worker nodes are not fault-tolerant?

A swarm manager nodes handles cluster management tasks such as: 1) Maintaining cluster state 2) Scheduling services 3) Serving swarm mode HTTP API endpoints You may execute any of the - docker ...
0
votes
1answer
228 views

what is the purpose of overlay network driver?

Below is the command run on ubuntu to create overlay network driver for swarm cluster instead of using bridge network driver $ docker network create -d overlay xyz $ docker network ls ...
0
votes
1answer
174 views

Performing maintenance on GFS2 inside pacemaker cluster

I will need to perform a maintenance on one of storage servers that provides GFS2 volume to a three node pacemaker cluster. Same cluster has an addition of 2 GFS2 volumes as well. Would it be safe to ...
1
vote
1answer
51 views

Run a shell script within an already running shell script

I have a small SLURM cluster, which I am using to run jobs. I run a script and while that script is running, I want to enter a specific key command "t + enter". I believe I should do it by submitting ...
1
vote
1answer
394 views

Slurm on desktop computers, how to prioritize the owner

In short: On a Slurm cluster, I need some computers to be available and responsive to their respective owners during work hours. Problem: I manage a small (but growing) heterogeneous cluster with ...
2
votes
1answer
1k views

PCS Stonith (fencing) will kill two node cluster if first is down

I have configured a two node physical server cluster (HP ProLiant DL560 Gen8) using pcs (corosync/pacemaker/pcsd). I have also configured fencing on them using fence_ilo4. The weird thing will happen ...
1
vote
1answer
268 views

Pacemaker: stopping pacemaker doesn't always move resources to other node

I have created a 2 node cluster on sles12. Following is the configuration - msnode1:~ # crm --version crm 3.0.0 msnode1:~ # corosync -v Corosync Cluster Engine, version '2.3.6' ...
1
vote
0answers
22 views

OpenLDAP cluster: deletion of an entry is not replicated

We have an OpenLDAP 2.4 cluster of three nodes configured in multi-master and accessed through a VIP in round-robin. The three machines run RHEL7. We noticed that deletion of an entry (done from a ...
0
votes
1answer
240 views

How to verify which is primary node in a HANA DB cluster operating on SLES 12?

Currently I have two HANA DB servers in a cluster and replication is not running. Upon checking from SUSE Hawk, node 2 is acting as primary node and node 1 is available. How do I confirm that node 2 ...
0
votes
0answers
95 views

rsync error code 23 with galera cluster

We are using Galera cluster with three nodes which syncs with rsync. All databases are synced correctly except one, for which I get the code 23 error. It syncs the schema but not the data. How can I ...
0
votes
0answers
1k views

Pacemaker cluster service can't start properly Couldn't mount filesystem

I have 3 nodes MySQL cluster on RHEL 7.5 with DRBD, Pacemaker and after start cluster getting following error. How can I overcome ? [root@db1 ~]# pcs status Cluster name: mysql_cluster Stack: ...
0
votes
2answers
573 views

Fencing of Redhat cluster 7 on VirtualBox

I am trying to create a 3 node Redhat Cluster 7 on Oracle VirtualBox VMs hosted on Windows Server 2012 R2. What agent can I use for fencing? How do I go about it? Note : I am creating a cluster for ...
2
votes
2answers
1k views

Expanding only certain variables inside a heredoc

This is an extension of the question I asked . Passing directory from command line to shell script I have a script which writes another script using a heredoc. I need to be able to write unexpanded ...
0
votes
0answers
442 views

Passing directory from command line to shell script [duplicate]

I want to pass directory as a command line argument to the shell script. #!/bin/bash sourcedir=/home/dir/data cd $sourcedir find "$PWD" -maxdepth 2 -name \*_R1*.fastq.gz > list1 fastq_list=$...
1
vote
1answer
139 views

restore $0 or $BASH_SOURCE after it is modified by the cluster

I am using a shared SLURM cluster. I am trying to get the path of the bash script from inside the script itself. There is already an excellent thread here: https://stackoverflow.com/questions/59895/...
0
votes
1answer
10 views

Redirect DB traffic dependent on type of request

I have a Postgres cluster managed by Patroni and want to be able to redirect DB requests to the replica if they are read requests and write requests to the master. Does anyone know of any tools that ...
1
vote
1answer
898 views

After failover Pacemaker moves resource back when node comes back

I'm using Pacemaker & Corosync for my cluster. When a node dies pacemaker moving my resources to another online node. Everything ok here. But when the dead node comes back, Pacemaker moving the ...
1
vote
1answer
898 views

PVS output not showing drive

I had created physical volume earlier i.e. /backup, /ndasdb, /processor, /latro. Don't know what exactly happened, but /ndasdb is not showing in any of this output result: Pvdisplay, Vgdisplay, and ...
0
votes
0answers
198 views

ORTE_ERROR_LOG, what does it mean?

1 [node18:02644] [[37701,0],0] ORTE_ERROR_LOG: File open failure in file ras_tm_module.c at line 142 2 [node18:02644] [[37701,0],0] ORTE_ERROR_LOG: File open failure in file ras_tm_module.c at line ...
6
votes
2answers
1k views

How do commands like fdisk -l find the sector size?

Does the MBR of the disk contain this information and therefore when i call a command like fdisk, a kernel level code eventually runs and reads it from a specific part in MBR? If so, which part of it? ...
1
vote
0answers
74 views

Distributed Block Device Cache for Linux

Is there an open source software available for Linux that implements a new block device based on an underlying block device (DRBD or shared SAN) and a distributed memory cache ? What I'm looking for ...
0
votes
1answer
681 views

Installing Percona XtraDB Cluster server

I want to Install the Percona XtraDB Cluster server package in Ubuntu 16.04.4 LTS but I got this error: sudo apt-get install percona-xtradb-cluster-57 Reading package lists... Done Building ...
2
votes
2answers
4k views

Memory Per Core [closed]

I am using the super computers in the network provided by Compute Canada and in the documentation page I see the following: I am quite curious - what is the concept of memory per core here? I thought ...
1
vote
1answer
43 views

Failover physical and virtual machine with replication. Is it possible?

I want to create HA cluster with 2 nodes. First node is physical machine, second is virtual (under ProxMox). Both running under Ubuntu. I know how to make HA with corosync and pacemaker. How i can ...
1
vote
1answer
41 views

NAT/Masquerade. Was working before, now is not

So for at least two months I had an IP tables configuration that was working on a compute cluster to provide internet access to compute nodes that were technically offline, via the headnode. We ...
13
votes
1answer
18k views

How to submit a job to a specific node using Slurm's sbatch command?

Our nodes are named node001 ... node0xx in our cluster. I wonder, is it possible to submit a job to a specific node using Slurm's sbatch command? If so, can someone post an example code for that?