Max's questions

On a Ubuntu 10.10 vm I installed PHP 5.3 via package manager (installed the whole lamp stack via tasksel install lamp) for quickly testing some project.

Now I need a newer version, PHP 5.4 beta, compiled from source. I downloaded the source, compiled and installed via configure --prefix=/usr && sudo make install

That worked for the cli version that is now using the PHP 5.4 binary, but as I expected it didnt replace the version apache is using, so apache still uses PHP 5.3.

How can I find out how the via tasksel installed version was compiled/what do I need to do in order to also upgrade the PHP version used with apache?

I made a rather simple startup script for my node.js program that should run on startup:

start on startup
stop on shutdown

script
    exec sudo -u max WEBSITES_DIR=/home/max/websites/ /usr/local/bin/node /home/max/websites/server.js 2>&1 >> /var/log/node.log
end script

If I use:

sudo start my-program 

it works, but when I reboot the machine I get the error message:

init: my-program main process (325) terminated with status 2

Anybody any idea on how to debug this?

Ok this is bit of a weird title, but the problem is weird, too.

We have a two server setup (www01 and www02) both running nginx and php as php-fcgi.

Nginx is configured to redirect the request to the other host if the php-fcgi process on the current host does not respond within 60 seconds.

What happened according to our logs: user is on www01 and posts a form to www01. The php-fcgi process does not respond and the request is forwarded by nginx to www02. php-fcgi again does not respond and the request is sent back to www01. Again php-fcgi does not respond and the request finally lands on www02 again. There it ends.

What is strange though: according to our database the form was indeed submitted AND the php application worked perfectly fine as the data submitted by the form was inserted 4 x (!) in the database each record created within a minute.

We couldnt find ANYTHING in the logs, except in the nginx error logs where nginx logged that it received upstream errors from www01/www02 as php-fcgi was not responding and forwarded the request to the other host. But no PHP error, no exception message, not even a warning.

How could this happen? Why did nginx think php-fcgi did not respond but the application ran anyway?

And how can we prevent this from happening again?

I managed to setup samba on a ubuntu server so that I can browse my home directory /home/max via windows.

Now I would like to share the directory /usr/local/apache2/htdocs - what do I have to do in order to share files between windows and the apache htdocs dir? The htdocs dir is owned by www-data.

I am pretty new to unix access permission stuff, so I really dont have a clue where to start with this. Any (easy) explanation is appreciated. Thanks!

In an upcoming project I have a two server setup: one is the application server and another, already existing runs the mysql server with databases I need to access. I contacted the server admin of the mysql server and the only way I can access the remote mysql databases is via "SSH tunnel". I have never done this before and never heard of it so far, so my question, are there any drawbacks, e. g. performance wise? Isnt it rather slow compared to directly accessing the mysql server on its default port?

In order to try out varnish for an upcoming project I installed it on an ubuntu server using this tutorial: http://varnish-cache.org/wiki/InstallationOnUbuntuDapper

The build process worked without any errors, but I cant start the varnish demon. I always get the error message

varnishd: error while loading shared libraries: libvarnish.so.1: cannot open shared object file: No such file or directory

But /usr/local/lib/libvarnish.so.1 clearly exists.

How can I tell varnish to look in that directory and load the library?

UPDATE

To answer the questions of cd34

    ldd `which varnishd`

outputs:

    linux-vdso.so.2 =>  (0x00007fff0a360000)
    libvarnish.so.1 => not found
    libvarnishcompat.so.1 => not found
    libvcl.so.1 => /usr/local/lib/libvcl.so.1 (0x00007f2a6fcaf000)
    libdl.so.2 => /lib/libdl.so.2 (0x00007f2a6faab000)
    libpthread.so.0 => /lib/libpthread.so.0 (0x00007f2a6f88f000)
    libnsl.so.1 => /lib/libnsl.so.1 (0x00007f2a6f675000)
    libm.so.6 => /lib/libm.so.6 (0x00007f2a6f3f1000)
    libc.so.6 => /lib/libc.so.6 (0x00007f2a6f082000)
    /lib64/ld-linux-x86-64.so.2 (0x00007f2a6fec7000)

Do you have varnish in two places on the machine, one from a previous attempt?
-> No, varnish is only installed once

Are you using 2.1.0 which was recently released?
Yes, I am using the most recent version

I´m planning for a large website that includes many static assets (js, css, images and thumbnails) in the generated pages. That website will use TYPO3 as CMS (is is a customer requirement).

I guess I could seriously improve performance / page load times by using a two server setup. One server where the main application (PHP) runs and another one where the static files sit being served by a trimmed down version of apache or something like lighthttpd.

Including e. g. js or css files from the file server is of course no big deal. Just use an absolute url http://static.example.com/js/main.js and be done with it.

But: that website will have pages with MANY thumbnails of e. g. product images on it. So I see two problems when the main application tries to create a thumbnail of some image:

1. the original image like products/some.jpg is uploaded on the static file server and therefore not on the same server as the PHP application which tries to create the thumbnail.

2. TYPO3 writes created thumbnails to a temp directory which is expected to be on the same server. Therefore, hundreds of thumbnails will be written and served from that temp directory which is on the same server as the main application -> the static file server is in that case basically useless, all thumbnails will be requested from the server of the main application.

So, my question is: how to overcome this shortcomings?

Is it possible to "symlink" some directories to another server?

So, for example, if PHP tries to open the original products image for thumbnail creation with imagecreate("products/some.jpg") the products folder actually "points" to the products folder on the static image server? I know something like this can be done with .htaccess but is it possible on file system level?

Eh, here is my problem: I am working in a webdevelopment agency (thats a problem but not the real problem, read on). Most of the time I choose the live server myself when creating a new website project. But now the customer already has a "server" (10 GB on a cheapo host!) and the "admin" refuses to give me ssh access to it. But I need to access the server via shell because many files will be transported (need to be able to upload and extract a tar) and I need to insert or create mysql dumps via command line. He argues FTP and phpmyadmin should be enough... as far as I know the webspace was just ordered to host the website, so no security critical apps are running there.

How can I either convince the admin to give me the ssh login or tell management that we need our own server? Anyone with similiar experiences?

This is really annoying as this is a very small project that should be done fast and now one has to fight in order to just get the work done...

I am working as a webdeveloper in a media company (around 25 employees) and I am not very happy with our current backup solution. I am no expert on this, but I am pretty sure the current setup is BS.

What is our current setup? Well, we have a fileserver with a 500 GB HD (sic!). If one remembers, one manually makes a .zip files and puts it on that server. Thats about it. Apart from the very small HD, nobody bothers making backups, maybe once a month. I think my boss has not backed up her data for maybe 3 months...

I would very much like to have an automated solution that creates a backup from all important files at night and... stores it somewhere.

The problem is: the backups can be quite large. As a developer I mainly work with text based stuff, so my backup including email etc. is around 7 GB. But the designers working in that company have backups in the size of 50 - 100 GB (already zipped). And there are at least 5 - 10 of these large backups.

Because of data loss recently I am pretty sure management would be willing to spend some $$$ on this if someone could come up with a good plan.

So, what would be a viable solution for a company of that size? Everbody in the company is working on OS X, if the operation system matters.

What I basically want to do in order to backup the website running on my webserver:

• create a dump of the mysql database
• copy / zip a certain folder containing the website in a directory
• zip the directory with the backuped website and the dump
• transfer it to another server or another HD or something like that

How can I automate this process to be executed every day at midnight?

Let me rephrase my question as I investigated further:

The problem:

I have a php script that is used to upload images on my windows webserver 2008. The files are created in the correct directory. The are created and owned by the user Network Service. Network Service has full access to the uploaded file.

As soon as I try to access the uploaded file (mostly an image) via HTTP, I get an 401.3 not authorized error.

Now, if I right-click on the not accessible image and grant IIS_IUSRS group read permissions via the security tab, the image can be accessed! By default IIS_IUSRS has NO access at all for the uploaded file.

The directory containing the image files has the correct access rights set. But each file that is new uploaded to the directory is permitted for IIS_IUSRS.

The question:

How can I grant IIS_IUSRS by default access to the newly uploaded file? The appPool of the website has its identity set to its default, I also tried setting it to "networkIdentity" or so, but that did not work either.

I am wondering what ports can / should be closed down on a windows webserver 2008. Port 80 of course not, and 3389 not, too, because I need a remote desktop connection.

netstat told me the server is listening on following ports:

  TCP    0.0.0.0:80             0.0.0.0:0              
  TCP    0.0.0.0:135            0.0.0.0:0              
  TCP    0.0.0.0:445            0.0.0.0:0              
  TCP    0.0.0.0:1311           0.0.0.0:0              
  TCP    0.0.0.0:3389           0.0.0.0:0              
  TCP    0.0.0.0:49152          0.0.0.0:0              
  TCP    0.0.0.0:49153          0.0.0.0:0              
  TCP    0.0.0.0:49154          0.0.0.0:0              
  TCP    0.0.0.0:49155          0.0.0.0:0              
  TCP    0.0.0.0:49156          0.0.0.0:0              
  TCP    0.0.0.0:49160          0.0.0.0:0              
  TCP    xxx.xxx.xxx.xxx:139       0.0.0.0:0           
  TCP    xxx.xxx.xxx.xxx:139       0.0.0.0:0           
  TCP    [::]:80                [::]:0                 
  TCP    [::]:135               [::]:0                 
  TCP    [::]:445               [::]:0                 
  TCP    [::]:1311              [::]:0                 
  TCP    [::]:3389              [::]:0                 
  TCP    [::]:49152             [::]:0                 
  TCP    [::]:49153             [::]:0                 
  TCP    [::]:49154             [::]:0                 
  TCP    [::]:49155             [::]:0                 
  TCP    [::]:49156             [::]:0                 
  TCP    [::]:49160             [::]:0                 

Anything that must not be active?

EDIT: Some clarifications:

What I want is a public webserver serving http (no https) requests and having remote desktop access to it. It would be no mail / exchange or intranet server.

I have an app running on a windows webserver 2008. It generates thumbnails which are then cached on disk.

But loading these static thumbnail images takes a bit long, as I dont have a slow connection. For example, Firebug tells me that most thumbnails (size between 30 - 50 kb) take between 600 ms and 1.2 second to load. And almost the whole time is spent while "waiting for response". Therefore I suspect that the server is responding too slow. Is there a way to optimize the serving of static content via IIS 7?

I managed to install the SMTP server on my windows webserver 2008 in order to send emails from PHP or .NET apps running on the server. Everything works fine, but now I am concerned about security. What do I need to do in order to make my SMTP installation safe? Any special settings? There are only one or two apps on the server all written by myself for the same client, so its not a hosting enviroment or something.

Here is what I did in order to set up the SMTP server: PHP mail() on windows webserver 2008

I have a Windows Webserver 2008 running with PHP installed. PHP works fine, except when I try to use the mail() function in oder to programmatically send emails via PHP. Then I get this error:

Warning: mail(): Failed to connect to mailserver at "localhost" port 25, verify your "SMTP" and "smtp_port" setting in php.ini or use ini_set() in ...

So, what do I have to install to confire in order to send emails via my PHP script?

I dont know much about mail configuration, but I have the bad feeling of being in for torture...

This is a bit of a follow up to one of my previous questions regarding how to install mysql on windows server 2008.

I managed to install the mysql server etc. without any problems, it runs, I can connect to it via PHP using "." as host (strange, but ok).

In order to follow best practices and not let mysql running with too much privilegs, I tried to assign the mysql service a special user.

So here is what I did:

Run this command to create a user:

net user mysql * /add /passwordreq:yes /passwordchg:no /comment:"runs MySQL services" (entering password information)

Run services.msc

Find mysql service in the list, and assign it the "mysql" user I created (in the "Logon" tab).

Windows tells me that changes are done, but the service must be restarted.

I restart mysql service and here comes the error - windows tells me that mysql service could not be started because it exited unexpected.

When I do not assign mysql a user, it works. Did I miss something?

What are some important points to consider when installing mysql 5.1 on windows webserver 2008?

I never installed mysql on any production windows server, so apart from not using the default user root with empty password - what else should be considered in order to safely install the mysql database?

I am using IIS as a webserver on my development machine for PHP webdevelopment. Or at least, I am trying to.

When there is a syntax error in a PHP script and I open that file in my webbrowser, I just get an 503 "internal server error" and the default IIS error page for this error. Some browsers dont open that file at all, possibly because of the 503 HTTP Response Header.

I would like IIS to act in that case just like the apache webserver: display the PHP file with the error anyway, so that the error message gets printed out.

How can this be done?

EDIT:

PHP settings:

display_errors

is on and

error_reporting

is set to E_ALL