Hortonworks says this: "Most often performance of a Hadoop cluster will not be constrained by disk speed – I/O and RAM limitations will be more important." *
How is disk speed not related to I/O limitations?
I have a problem when I try to build from my spec file. I run this:
rpmbuild -ba newpackage.spec
I see a great deal of output. It ends with these lines:
configure: error: Cannot find Lightning MDB
error: Bad exit status from /var/tmp/rpm-tmp.dKgUSh (%build)
RPM build errors:
Bad exit status from /var/tmp/rpm-tmp.dKgUSh (%build)
I tried to install lmdb-devel package to address the "Cannot find Lightning MDB." There were many dependency packages. I am using RHEL 7. I couldn't even find where to download all the dependencies. What should I do? I'm concerned that installing the lmdb-devel won't even help me anyway.
Here is what my newpackage.spec file looks like:
Name: cfengine
Version: 1
Release: 1%{?dist}
Summary: First time
License: GNU
#URL:
Source0: http://s3.amazonaws.com/cfengine.package-repos/tarballs/cfengine-3.6.5.tar.gz
BuildRequires: gettext
# Requires:
%description
This is a test.
%prep
%setup -n cfengine-3.6.5
%build
%configure
make %{?_smp_mflags}
%install
rm -rf $RPM_BUILD_ROOT
%make_install
%files
%doc
%changelog
I installed rpmlint and rpmdevtools. I used rpmdevtools to create a template of a .spec file. I modified the spec file so it will install cfengine. I ran rpmlint on the spec file, and I saw no warnings or errors. I then tried to run the specfile for real, but it failed. Here is the command I ran:
$ rpmbuild -ba newpackage.spec
Here is what I saw:
Executing(%prep): /bin/sh -e /var/tmp/rpm-tmp.wNGgcV
+ umask 022
+ cd /home/ec2-user//BUILD
+ cd /home/ec2-user/BUILD
+ rm -rf Propulsion-1
+ /usr/bin/tar -xf -
+ /usr/bin/gzip -dc /home/ec2-user/SOURCES/cfengine-3.6.5.tar.gz
+ STATUS=0
+ '[' 0 -ne 0 ']'
+ cd Propulsion-1
/var/tmp/rpm-tmp.wNGgcV: line 35: cd: Propulsion-1: No such file or directory
error: Bad exit status from /var/tmp/rpm-tmp.wNGgcV (%prep)
RPM build errors:
Bad exit status from /var/tmp/rpm-tmp.wNGgcV (%prep)
The file above on line 35 has this:
cd 'Propulsion-1'
The "Propulsion" name comes from the .spec file. Evidently, the single quotes, hyphen and 1 come from the rpm-build command processing.
What should I do to not have an RPM build error? I want my spec file to work for the installation of cfengine.
Here is the spec file:
Name: Propulsion
Version: 1
Release: 1%{?dist}
Summary: First time
License: GNU
#URL:
Source0: http://s3.amazonaws.com/cfengine.package-repos/tarballs/cfengine-3.6.5.tar.gz
BuildRequires: gettext
# Requires:
%description
This is a test.
%prep
%setup -q
%build
%configure
make %{?_smp_mflags}
%install
rm -rf $RPM_BUILD_ROOT
%make_install
%files
%doc
%changelog
In the AWS management console, I stopped a Linux instance. Then I restarted it. I tried to reconnect via Putty. My putty session has an "Auto-login username" set and a private key for SSH authentication set. AWS's DHCP gave my instance a new IP address upon rebooting. So I changed the IP address in Putty. When I try to connect via Putty I get this error:
"Disconnected: No supported authentication methods available (server sent: publickey,gssapi-keyex,gssapi-with-mic)"
How do I connect via Putty to my Linux instance? I was able to connect fine before I rebooted it using AWS. I tried using a different user. But I don't even get to the password prompt. It says "Server refused our key."
To install Puppet Master on RedHat Linux version 7, you need to install an optional channel. Can someone provide directions on how to do this? I want to install Puppet Master. I tried to install Puppet Master without the optional channel. The main thing that is wrong is that I cannot start puppetmaster.
I run this: /etc/init.d/puppetmaster restart
But I get this: -bash: /etc/init.d/puppetmaster: No such file or directory
The Puppet Labs website has a broken link for installing the optional channel. There must be directions somewhere else.
I'm using RedHat version 7. My AWS instances are in a security group with ample permissions for inbound connectivity. I can do Yum updates. The servers can ping each other. I can transfer files from my workstation to the servers. I even temporarily allowed all traffic from any IP address. This didn't seem to help me.
I understand that RedHat v7 doesn't use IP tables by default. RedHat v.7 uses firewalld or something similar. I tried shutting off the firewall.
I used this command to configure the firewall: system-config-firewall-tui I unchecked the "Enabled" option for the firewall and saved changes.
I ran these commands: systemctl stop iptables systemctl stop firewalld systemctl disable firewalld
I have root permissions. The nmap utility indicates that almost every port is blocked. I tried the nmap utility against specific ports and multiple ports from one Linux server to itself and a second Linux server on the network to the original Linux server. I would get something like this:
Host is up (0.00042s latency). PORT STATE SERVICE VERSION 6379/tcp closed unknown
I double checked the nmap results with a bash script with these lines:
exec 6<>/dev/tcp/127.0.0.1/6379 || echo "Not listening on port 6379" exec 6>&- # close output connection exec 6<&- # close input connection
The above bash script indicated that the port was closed. (The script was from this link: https://stackoverflow.com/questions/9609130/quick-way-to-find-if-a-port-is-open-on-linux)
I enabled the iptables just to see if a firewall needed to be running. I ran this to try to open port 6379:
iptables -A INPUT -p tcp --dport 6379 -j ACCEPT
The above command has no response. I just go to another prompt.
What can I do to open port 6379? It seems like only port 22 is open.