I had downloaded cygwin from one mirror site, and after a few months when I try to update now I get the message that the "current ini file is from a newer version of setup-x86_64.exe"
If I download a new ini file, does it also mean that the entire cygwin installation will have to be downloaded again? That will take a huge time and I would like to avoid it.
Also, what is a stable mirror site for cygwin? On some update occassions, I have got the message that the site is no more a "recommended" site, and I should select another mirror site.
The answer to your main question is NO. That message appears because the setup-x86_64.exe file has been updated. You need to download a recent version at http://www.cygwin.com/setup-x86_64.exe.
Mirror sites are updated on a schedule, so the updates at mirrors are always behind the main Cygwin site by a certain amount. Some sites update more frequently than others. From the Cygwin mirror admin site:
Given the granularity of checking, it is possible that your site will
be polled after a package update on cygwin.com but before your site
has pulled the update. So, it is expected that from, time-to-time,
your site may fall off the mirror list temporarily. If your mirror
site has not been listed for a day or so that means that some cygwin
packages on your site are not current. Check the cygwin-announce
archives for announcements about recent package updates and ensure
that your site has those packages. Once your site has the recent
packages it will be re-added automatically.
and
So, if your site was dropped from the list it means that a program has
determined that the files on your site are not up-to-date. Unless your
site has been out of date for (currently) 100 runs of the program, it
will be re-added automatically when it becomes current. So, if you see
that your site has been dropped from the mirrors list do not panic.
I’ve generally found that unless you repeatedly get the message about it not being a recommended site, there's nothing to worry about, just try again in a few hours or the next day. Picking the site closest to you is generally a good bet.
You just need to download a newer version of the setup.exe file and perform the update with that.
During the setup you'll be presented with a list of recommended mirror sites, choose one near to you and you're good to go.
No worries. :-)
Related
As a newbie with phpBB forum (3.1), I understand that modes may change data in forum database and files.
Therefore, I would like to backup the site before mode installation, so that the previous state could be easily restores in case it is necessary.
What are the exact instructions to install and de-install modes safely?
I'm guessing you mean 'Mods' instead of 'modes'?
phpBB v3.1.x does not use Mods like the old 3.0.x branch where core code needed to be changed. Instead it uses extensions which are uploaded to your server and simply activated in your ACP.
As a precaution it is always advised to back up your database before making any changes to your site.
Full instructions are on the phpBB sie here - https://www.phpbb.com/extensions/installing/
How can I create a CentOS 5.8 .iso image with custom packages? I have to create an iso with only the packages needed for our production system. I already have all the rpms with their dependencies resolved in a folder. I have successfully created a repository from that folder with createrepo.
As I understand it, I should put the rpm files in the CentOS folder, and repodata folder should contain the metadata needed for a package manager. I don't know if I should modify the existing comps.xml file or create a new one, or which structure to use, since this is only a subset of packages contained on a default CentOS installation disk.
I know it is probably futile to delete packages from the default iso, but that is my work order and there's not much to be done there. (There are also some packages not available in the default iso)
Much appreciated
I feel as though your question is a bit vague for a topic with this sort of breadth but I'll do my best to offer an answer. I think you should use Kickstart for this task as it's going to result in a much happier customer whether they're internal or external, and easier management for you going forward when things get updated. Start by reviewing the CentOS documentation, if you are already this far and just asking about removing packages, check out this section of the docs, it talks about specifying your packages and removing the ones you don't want.
If you only have the one style of production machine, then this should take care of it. if you have multiple different configurations I'd suggest taking a look at a configuration management tool such as Ansible, Puppet, or Salt. This would allow you to provide a base image via Kickstart, then build off of that image depending on the needs of whoever is consuming the system.
I have live website running on MODx Revolution 2.1.3pl. Some days back I had to restore my entire site from backup. This messed up some file ownerships (for packages installed and images uploaded etc.) because in my server PHP runs as 'nobody' user which is different from my cPanel user.
Now I can't change much things on the server(like installing suPHP because its a shared server) and I don't know which all files are created by PHP, I decided to wipe the site clean and perform a clean install. My site has a large number of already published resources which is impossible to be posted into the new site individually.
Is there any way that I can transfer those resources to the new installation?
Why don't you create a mysql dump of your old site (with phpmyadmin or the like) and import this into a new database, which you use to run your new site from?
I've not tried it myself but provisioner seems to do (or at least claim to) what you need.
I am using Installshield 2010. I made a basic MSI setup, and I am having a problem with uninstalling.
Uninstalling does not remove all folders.
After a complete install, I then uninstall only to find that there is still a path leftover.
For example, C:\ProgramFiles\CompanyName\Account\User_1234.xml
still remains, while everything else is gone.
The User_1234.xml file is generated by the application, and this file is not normally in the install process. So it makes sense why this file may have some problems getting rid of it (installer side).
At the same time, I would think that it would be removed due to the fact that it is within the CompanyName folder.
Is there any way to specify to get rid of everything inside and including the CompanyName folder?
Thanks for any ideas.
You have a couple problems and I'll address the obvious one first. Yes, Windows Installer default behavior is to not delete user data. This could be files not installed by the installer or files that were modified since the installer modified them. To get the installer to remove them, you must author the RemoveFiles table to teach it which files to remove.
Remove Files Table (Windows Installer)
Now I'll address the less obvious problem. You shouldn't expect your application to be able to write to ProgramFiles at runtime. I don't know what your software does but you should seriously consider this aspect of your application design.
I'd like to download the Trac database so I can view its tickets offline. Is there anyway to achieve this? I.e. if I need to leave the office and bring my laptop with me, how can I bring the tickets with me without having to connect to the company network?
I know that Mylyn can download and sync tickets via it's trac connector but I'd like some stand-alone viewer.
See Simple Defects (SD).
I particularly like the "One-tweet install" idea.
I’m installing #SD (http://syncwith.us)
after reading about it on #StackOverflow
curl fsck.com/sd|perl;
export $PATH=~/sd/bin:$PATH; sd
Note that you can clone Trac (and other bugtrackers) in SD:
sd clone --from trac:https://trac.parrot.org/parrot
Seeing as you don't want to install a server, how about using RSS? IIRC, Trac let lets you get RSS feeds for each person, so you can have a feed of things assigned to you.
All you need do then is get a nice client that will download these tickets. You should be able to access a plaintext version without internet connection.
If that's not flexible enough, you could write a script on the server to publish a feed using the database directly.
And if RSS isn't for you (and your email is available offline), you could mail reports home. Trac also has this built in.
The default Trac installation uses a combination of SQLite to matintain all of the data. Attachements are stored on the file system.
In the folder containing the trac site, find \db\trac.db
This file can be viewed using the SQLite manager Firefox Addon
Happy hunting.
And if RSS or email isn't your notification of choice, there's a trac plugin that will let you receive task notifications on your Remember The Milk todo list.
See: http://1.www.rememberthemilk.com/forums/ideas/3580/?forum=ideas&hl=bs&topic=3580
If your objective is simply to view the tickets offline, how about
Run a report with all the tickets (or all those you're interested in).
Select either the comma-delimited or tab-delimited download link at the bottom of the page.
Import the downloaded file into Excel.
you could install it on a local machine
You can host the trac locally and set up the connectionstring point to your dowloaded database.
Sure. Install a web server locally, install trac, get it set up the same (or similar) way to the way it is on the live version and then script the server to publish db backups and write a local script to download those and restore them over your database.
It's not simple (installing Trac is a battle on its own from my experience of it) but every element is highly googleable =)
The trac client FatBug (http://fat-bug.com/) listed in
https://trac.edgewall.org/wiki/Clients
seems to do the exact what was described by the OP. I bumped into it after I just checked SD. SD seems trival on Linux, but heavy on Windows, it depends on Perl & CPAN.