Due to some situations, only one server (main server) can access internet and others connect to this server using vlan. So I want to create a repo in this server to install the following rpm:
http://downloads.eucalyptus.com/software/eucalyptus/4.1/centos/6/x86_64/eucalyptus-release-4.1.el6.noarch.rpm
I installed vsftpd on my main server and download the rpm file to the ftp folder, then run the following to createrepo:
createrepo /var/ftp/pub/localrepo/
Then I added the repo to my client server, but the following error exists when I install the eucalyptus package:
No package eucalyptus-cc available.
Any ideas?
Related
I'm fairly new to Linux and am running CentOS 7. I can do the basics of navigating, creating dirctories, permissions, unzip applications and make SL to the apps but I have no idea what to do with an "rpm" file.
The system that I am trying to install MongoDB 4.2.x on does not have an outside connection. I have the individual rpm files for MongoDB but not the tgz which some article I read said I should use.
There is no way for me to get the tgz over to the machine unless I wait several days.
With only the .rpm files, how do I install MongoDB?
[https://docs.mongodb.com/manual/tutorial/install-mongodb-on-red-hat/][1]
The instructions in the above link say to create the .repo file but that makes a reference to the external URL which the system does not have access to.
[mongodb-org-4.2]
name=MongoDB Repository
baseurl=https://repo.mongodb.org/yum/redhat/$releasever/mongodb-org/4.2/x86_64/
*
Again, with only the .rpm files, how do I install MongoDB?
*
You do need some kind of access to the outside world; you need to get the rpm files on your machine...
you can just manually download the rpm files you need from https://repo.mongodb.org/yum/redhat/7/mongodb-org/4.2/x86_64/RPMS/, copy them to your machine, and then run
yum install /path/to/downloaded/mongodb*rpm
Being a total newbie, I want to install Odoo on my CentOs device via cloning github repository. However, I am not certain about the installation directory I should choose. As you can see, the ftp directory is empty, shall I install it there?
The installation folder is not too important as long as the permissions are opened for the user that will run the server.
Usually, it's installed under /opt/odoo
sudo mkdir /opt/odoo
cd /opt/odoo
sudo git clone https://github.com/odoo/odoo.git
I have created a package which I can install locally:
cinst <packageName> -source "C:\Host\<packageName>"
But when I put this package on a file server served over http from a Virtual Directory by IIS, it fails.
cinst <packageName> -source "http://server/packages/<packageName>"
The target machine can browse to the package directory in a browser, so it is able to see the server, the folder and has list and read permissions.
The error
[ERROR] - install not installed. The package was not found with the source(s) listed.
Fiddler shows the requests:
http://server/packages/<packageName>
GET /packages/<packageName>
301 redirect to http://server:port/packages/<packageName>
http://server:port/packages/<packageName>
200 with <packageName>.nuspec and .nupkg files in the response
then
GET /packages/<packageName>/$metadata
404 File or directory not found
My two thoughts:
Is it a Windows security measure stopping me
Is my package missing some metadata
I used to managing packages on *nix and I have only recently started to work in a Windows environment, so feel free to state anything obvious!
Environment:
Information.PlatformVersion='6.1.7601.65536'|
Information.PlatformName='Windows 7'|
Information.ChocolateyVersion='0.9.9.11'|
No additional pointers in the chocolatey log with the -debug set
Not using an IDE or a Repo, all just simple.
TY!
Unfortunately NuGet (and by proxy Chocolatey) does not work with bare http. It requires OData right now, although in the future the Chocolatey team plans to allow working with just http indexes.
Here's a resource explaining your options for package hosting: https://github.com/chocolatey/choco/wiki/How-To-Host-Feed
I tried to install Phalcon on CentOS 6.5. This is what I did:
Edit etc/yum.conf: remove the line ‘exclude=’ (temporarily)
Terminal:
Sudo yum update (to update everything on the system)
Yum install php-devel
Sudo yum install git
git clone git://github.com/phalcon/cphalcon.git
cd cphalcon/build
sudo ./install
Add a file called phalcon.ini in /etc/php.d/ with this content:
extension=phalcon.so
Put back the original etc/yum.conf file
Reboot server.
Before I rebooted the server, "Apache is functioning normally" was displayed when I typed the IP address in my browser. Now when I got there, I get a browser message about being unable to establish a connection. When I type :2222 after the ip address, I can get access to directadmin.
What did I do wrong? (this is the second time I tried to reinstall it completely...)
Thank you very much
Phalcon should be loaded after all other PHP extensions are loaded. This has to do with the patches used by CentOS: they change the way the extensions are loaded into PHP.
I would remove extension=phalcon.so from php.ini and create /etc/php.d/zz-phalcon.ini with
extension=phalcon.so
and restart the web server.
EDIT: I explained this in details in the official forum.
You can't properly install Phalcon when using DirectAdmin.
I'd like to install a PHP extension on a bluehost shared site; specifically the MongoDB driver. Since pecl is unable to write to the primary server directory that has all the installed extensions, I'd like to install the mongo.so file to a directory I specify under my home. The closest article I found on the web was:
http://www.site5.com/blog/programming/php/how-to-setup-your-own-php-pear-environment/20090624/
However, after following the steps when I use the "pecl install mongo" command, it still keeps trying to install to bluehost's central directory on the server.
According to my web host's technical support team, utilising the pecl installer attempts to install the extension server-wide rather than under your account only. My web host doesn't allow server-wide installations in their shared environment for security reasons and because they want to keep their fleet universally the same across the board. I suspect your host is the same.
However, they did suggest I download, configure and install the pecl package (pecl_http) in my account only (rather than server-wide) via the following manual process:
$ cd ~/
$ wget http://pecl.php.net/get/pecl_http
$ tar -zxvf pecl_http.tar.gz
$ cd pecl_http
$ phpize
$ ./configure --prefix=~/bin
$ make
$ make test
$ make install
A successful install will have created extname.so and put it into the
PHP extensions directory. You'll need to edit php.ini (in my case,
this was a copy of the standard php.ini file that I placed in the same
folder as the script using the extension) and add an
extension=extname.so line (in my case, extension=http.so) before you can use the extension.
http://www.php.net/manual/en/install.pecl.phpize.php
Note that the tilde character (~) in the above code refers to the home directory of the current user (e.g. /home/username on my host's server).
Issues you may run into
When using the tar command, a "cannot open: not a directory" error
appeared as pecl_http had been downloaded without a file extension.
This was easily corrected:
mv pecl_http pecl_http.tar.gz
When using the make install command, a "cp: cannot create regular
file...: Permission denied" errror appeared. I resolved this issue
by changing the ext_dir for pecl...
pecl config-set ext_dir /home/username/bin/extensions
...and re-running make install. I also had to copy the http.so extension to /home/username/bin/extensions and then reference that location in my php.ini file:
extension_dir = "/home/username/bin/extensions"
this sounds like you don't have root access to your server. if your need to compile anything you must be have root access permission to server, or maybe you must be one of the sudoers.