Discuss

discuss@lists.centos.org

34758 discussions

Re: [CentOS] Ruby rails rpm package
by James B. Byrne 21 Nov '12

21 Nov '12

On Tue, November 20, 2012 06:53, C. L. Martinez wrote: > On Tue, Nov 20, 2012 at 11:23 AM, Phil Dobbin <bukowskiscat(a)gmail.com> > wrote: >> On 11/20/2012 08:39 AM, C. L. Martinez wrote: >> >>> Hi all, >>> >>> Somebody knows if exists some rpm package for ruby rails?? >> >> Normally, rubygems is the way to go to install Rails. >> >> Cheers, >> >> Phil... >> > > Sure, but how can I update rubygems installed in one system?? > > The main problem with RHEL and Ruby on Rails is that the version of Ruby available for EL6 from reputable repositories is too old. RoR 3.2 can get by on 1.8.7 but support for Ruby versions prior to 1.9.3 is being dropped in the forthcoming RoR v4 release. Thus, a RHEL/CentOS rpm package for Rails really will not give you much, if anything. I suggest that you investigate both RBENV and RVM as alternatives to using the rpm packaged Ruby. I advise you consider that for RoR projects the Bundler gem is the preferred way of installing and managing project specific packages, including Rails itself. The down side to this approach is that your production servers need to have development tools installed to build the Ruby interpreter and the support gems. The up side is that you can version specific ruby vms and gemsets on a project by project basis. https://github.com/sstephenson/rbenv https://rvm.io/ http://gembundler.com/ I used rvm almost from its inception but have recently changed to rbenv as this has a much smaller footprint on the user's environment. But both are excellent products. If you are installing RoR for a production environment then you will almost certainly need to consider using Passenger (mod-rails) as well. https://www.phusionpassenger.com/ HTH. -- *** E-Mail is NOT a SECURE channel *** James B. Byrne mailto:ByrneJB@Harte-Lyne.ca Harte & Lyne Limited http://www.harte-lyne.ca 9 Brockley Drive vox: +1 905 561 1241 Hamilton, Ontario fax: +1 905 561 0757 Canada L8E 3C3

3 3

Re: [CentOS] Ruby rails rpm package
by James B. Byrne 21 Nov '12

21 Nov '12

On Tue, November 20, 2012 10:28, m.roth(a)5-cent.us wrote: > > Here, they use Ruby, the enterprise version - is that what you mean by > RBENV or RVM? The next release of ruby? from RH? will be the 1.93 or > some such, and include all the stuff in the enterprise version. RNENV and RVM are Ruby language installers that create a custom programming environment by user. REE is simply a modified Ruby VM based on 1.8.7 which was produced by the same people who provide the Passenger gem. Support of REE is / will be discontinued as MRI (Matz Ruby Interpreter) version 1.9.3 addresses most of the issues with v 1.8 that REE attempted to correct. > Development tools on a production box are a very, VERY bad idea. I > assume you can build the ruby app on your development box, and then > move it as a package to test, then prod. I am not sure that I can agree with this. All of our application servers are sealed. There are no local users other than administrators and application pseudo-users so the presence or absence of development tools is mostly moot. Anyone who penetrates a sealed server already knows enough that the absence of development tools is a negligible inconvenience. On the other hand, many RubyGems must be custom built on their target hosts. This is particularly true for DB adapters. To keep these gems up-to-date requires build tools. I suppose that I could direct that all build tools be removed after a gem update and reinstalled as required but the illusionary security benefit is simply not worth the very real labour cost. > > I've also seen an article or two about ruby not scaling up well. > From my experience here, the apps seem to be *very* fragile, and First, there are a lot of fragile COBOL applications that have been running in large corporations for decades. I know, I worked on some of them. On one system, run by a now defunct telecommunications giant, the corporate accounting staff had to calculate foreign currency adjustments off line and explicitly ADD the resulting figures to the production cost system as component items. This was necessary because the existing system did not handle foreign currencies and NO ONE was going to take the responsibility for breaking the entire cost accounting application just to add that feature. Just consider how fragile that is. The reason that there are a lot of fragile Ruby (and a plethora of any other languages that you care to name) Apps is mostly due to the sudden popularity of the language generated by the creation of a 'cool' new framework (in this case Ruby on Rails). The result is attraction of swarms of people from other languages who throw up (in both senses of the term) hastily built, generally untested, toy applications that simply are not designed to handle growth; or simply not designed at all. That is likely the case with most of these supposed 'fragile' apps. Apps which no-one seems able to name however. If one comes to Ruby from Python or PHP or MVB or dot.net or Java or C++ or C and just starts coding then one likely will end up with a Python, PHP, VB, dot.net, Java, C++ or C application. It will be written in Ruby but it will not be a Ruby app. The problem is succinctly put by the observation that one can write Fortran in any language. Of course, if you come from Smalltalk, then the case is somewhat altered. But, I digress. > it reminds me of python 10-12 years ago, where updating it one or > two subreleases broke everything that had been working, including > system tools. I cannot speak to what happened with Python a dozen years ago but it seems to me irrelevant to what is happening with Ruby today. > > ObStmt: No, I don't like ruby. I would never have guessed. -- *** E-Mail is NOT a SECURE channel *** James B. Byrne mailto:ByrneJB@Harte-Lyne.ca Harte & Lyne Limited http://www.harte-lyne.ca 9 Brockley Drive vox: +1 905 561 1241 Hamilton, Ontario fax: +1 905 561 0757 Canada L8E 3C3

1 0

selinux policy and httpd
by m.roth＠5-cent.us 21 Nov '12

21 Nov '12

I upgraded a development server last week, and it started spewing selinux errors to the log. I googled. What finally *seems* to have stopped it was a) setsebool -P httpd_setrlimit 1 b) yum downgrade selinux-policy\* This is on a 6.3 box. Has anyone else seen this behaviour? mark

5 7

HP Data protector problem on Centos 6.3
by Todor Petkov 21 Nov '12

21 Nov '12

Hello, has anyone tried out HP Data Protector 7.0 on Centos 6.3? I have problems with uiproxyd, similar to http://h30499.www3.hp.com/t5/Data-Protector-Support-and-News/DP-7-and-RedHa… It fails to start, and provides no usable information, neither with debug or strace. Thanks in advance,

1 0

System->Administation->Network
by Steve Campbell 21 Nov '12

21 Nov '12

I found quite a bit on google about the problem that "Network" does not show up under the System->Administration menu any more. All the solutions seem to indicate that I just need to install system-config-network-tui, but that doesn't change anything. I've turn off Network Manager before trying to do any of this because I want to configure my network offline. A little push in the right direction would be appreciated. Thanks steve campbell

6 8

"upstream" Storage Server fully OSS?
by Ted Miller 21 Nov '12

21 Nov '12

Is the "upstream" Storage Server fully open source, or are parts of it closed source? Are the RPMs to build one already in the Centos repo? If not, are there any plans to offer them? I am looking for something free to use in Haiti, that will offer redundant file storage and automatic failover to a second set of hardware in case of a failure of the primary hardware. If RHSS is not available or suitable, other suggestions welcome. I need a file system/server with: * primary function is serving MP3 files for playback in a radio station environment in Haiti * if the system goes down all your clients (listeners) know it * they know it NOW * they know how long it takes to get it back up * High Availability as the primary concern * ability to administrate via web interface or similar by non-Linux-savvy IT staff. * ability to grow file system from 2-3TB to 20-50TB by simply adding disks and/or adding 'bricks' * clients will all be Windows computers, so files accessible by CIFS * critical application is read-only * prefer a system that would continue serving files even if the network goes down (but have not found such a system yet for Windows clients). This would require something like Ceph with a full (non-server) windows client, so the local node would continue to function until the network came back up. * throughput is not a large issue Ted Miller Elkhart, IN

7 10

How can I prevent apache from mounting all home directories?
by Alfred von Campe 20 Nov '12

20 Nov '12

We use automounted user home directories on our CentOS 6.3 desktops, and on the desktops on which we start apache, I have noticed that all 900+ home directories listed in NIS get (and stay!) mounted. If I don't start apache, this doesn't happen. We don't need access to home directories from apache, so I would like to prevent apache from mounting these home directories, but I haven't been able to figure out how to do it. We use the stock httpd.conf file, and I commented out the "LoadModule userdir_module" line to no avail. Any other suggestions? Alfred

5 15

Ruby rails rpm package
by C. L. Martinez 20 Nov '12

20 Nov '12

Hi all, Somebody knows if exists some rpm package for ruby rails?? Thanks.

2 4

SAMBA and Active Directory Intergration Problem
by Bonnie B Mtengwa 20 Nov '12

20 Nov '12

I have a CentOS6.3 SAMBA file server. All the users are on a Windows Server 2008 Active Directory Domain Controller and authentication is done using Winbind integration on SAMBA . Everything works well except that if someone logs on to the Samba Server, they see all the shared folders there and also a folder with their own name %USER%. This Folder is not accessible, one cannot enter into this folder, I have noticed that it's not being automatically created in the home folder : i.e /home/%USER%/ is not there. My smb.conf is as follows workgroup = Domain0 security = ads realm = Domain.com encrypt passwords = yes preferred master = no template shell = /bin/bash template homedir = /home/%D/%U winbind separator = $ ; template primary group = "Domain Users" winbind uid = 600-20000 winbind gid = 600-20000 winbind use default domain = yes winbind nested groups = yes winbind enum users = yes winbind enum groups = yes enhanced browsing = yes winbind offline logon = false password server = dc.domain.com [homes] create mask = 0775 directory mask = 0775 browseable = no comment = Home Directories valid users = %U writable = yes anyone who can help me on this? Bonnie

3 2

Re: [CentOS] "upstream" Storage Server fully OSS?
by Ted Miller 20 Nov '12

20 Nov '12

> On 2012-11-19 9:48 PM, Ted Miller wrote: >> Is the "upstream" Storage Server fully open source, or are parts of it >> closed source? >> >> Are the RPMs to build one already in the Centos repo? If not, are there >> any plans to offer them? >> >> I am looking for something free to use in Haiti, that will offer redundant >> file storage and automatic failover to a second set of hardware in case of >> a failure of the primary hardware. >> >> If RHSS is not available or suitable, other suggestions welcome. I need a >> file system/server with: >> >> * primary function is serving MP3 files for playback in a radio station >> environment in Haiti >> * if the system goes down all your clients (listeners) know it >> * they know it NOW >> * they know how long it takes to get it back up >> * High Availability as the primary concern >> * ability to administrate via web interface or similar by non-Linux-savvy >> IT staff. >> * ability to grow file system from 2-3TB to 20-50TB by simply adding disks >> and/or adding 'bricks' >> * clients will all be Windows computers, so files accessible by CIFS >> * critical application is read-only >> * prefer a system that would continue serving files even if the network >> goes down (but have not found such a system yet for Windows clients). This >> would require something like Ceph with a full (non-server) windows client, >> so the local node would continue to function until the network came back up. >> * throughput is not a large issue >> >> Ted Miller >> Elkhart, IN >> >> _______________________________________________ >> CentOS mailing list >> CentOS(a)centos.org >> http://lists.centos.org/mailman/listinfo/centos > On 11/19/2012 09:43 AM, Banyan He wrote: > What is this upstream you talk about? Do you have a website for it? > Kinda interest into this thing. Unfortunantely, I dont find it from > google. > ------------ > Banyan He > Blog: http://www.rootong.com > Email: banyan(a)rootong.com "upstream (noun) In free and open source projects, the upstream of a program or set of programs is the project that develops those programs." From Fedora (https://fedoraproject.org/wiki/Staying_close_to_upstream_projects) In general, when Centos folks refer to upstream in general terms, they are talking about RedHat. It often seems to be used in order to avoid too much use of the Trademarked name that makes Centos possible. If someone is talking about a specific program (say apache or MySQL) and they refer to upstream, they are probably referring to the developers of that specific program. Ted Miller P.S. I moved your top-post to the bottom of the email, in order to conform with list protocol. Learn to bottom-post. It makes it easier for other readers to follow a thread as it progresses.

1 0

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Discuss