Luke's questions

I have a multi VM Windows (Server 2019) solution, hosted in a datacentre, with services which require a Redis cache. Due to security concerns, the Windows fork of Redis is not allowable. Therefore, I have a Redis installation on a RHEL 7 server.

We have a network share on one of our Windows machines which contains various configuration files for the solution, as well as backups. It is to this share that I configured Redis to write out its dump.rdb file every 20 minutes (or so many changes), using a CIFS mounted share with service user AD domain credentials (having administrative privileges on the share server).

This set up had been working perfectly for about 2 months. Last night around midnight the Redis backup started failing. Permissions denied on the dump.rdb file. On investigation, it seems the ACL permissions in Windows cannot be viewed (You Must Have Read Permissions to view the Properties of This Object), and operations (Rename, Copy, Move, Delete etc.) all fail with access denied - even from an elevated command prompt. On Linux, ls -l shows full rwx permissions for all users, but any attempt to access the file returns an error saying that no such file exists.

So there is nothing we can do with this file. We cannot get rid of it. For the time being, we have told Redis to dump to a different folder, but who knows whether/when that file will get corrupted permissions too.

Attempts to google the issue have come up with nothing which is relevant to our specific issue. Is there anything we can do to remove the dump file? Something to override permissions? Also, is there any known issue with CIFS that can cause this kind of error? Perhaps I have things configured incorrectly. Here is my cifs mount entry in the fstab file:

//<ServerIP>/DataShare /mnt/DataShare cifs _netdev,credentials=/root/.credfile,dir_mode=0777,file_mode=0777 0 0

The .credfile is as follows:

username=svc_user
password=<password>
domain=mydomain

So I have a large batch file, and in the middle it hits a start /wait command to run a silent install of an MSI. I'd like to have some animation or something to indicate that the batch file is in fact doing something, as it can take some minutes. Is there any way to fire off the start /wait command but still keep some looping code in the batch file to run an animation that can show it is working on something?

I have a service in the form of a node.js application set up with Systemd on Raspbian Jessie and it is using its own user account. However, I am finding that the service does not run correctly because it does not have the necessary permissions. One of the node modules I installed requires root access. If I run the application manually with sudo everything works fine.

Is there a way to tell systemd to run the service with sudo?

I have a problem with multiple processes needing to bind a TCP socket to the same port in Windows Server 2008.

At the moment I believe this is causing some unexpected erratic behavior as a race condition between processes exists, where you can't know which packet will be delivered to which process.

Separating the processes to different machines is apparently out of the question, so one solution I would like to try is binding on this port through different IP addresses.

Now, I know I can use netsh interface ipv4 add address to add static IPs in Windows. However, is it possible to add addresses and allow the DHCP to assign IPs to each of these virtual interfaces?

So, I have two web applications on one server. I have a main one which is served when the regular domain is used, but I want a subdomain to point to the other app.

I have set up virtual hosts in the way that I have determined they should be with what I have found in google searches, but it still doesn't work. The subdomain just serves up the main app. Here is the contents of my config file.

<VirtualHost *:80>
        ServerName www.luketimoth.me
        DocumentRoot /var/www/luketimothy/public

       <Directory /var/www/luketimothy/public>
          <IfModule mod_rewrite.c>
          Options -MultiViews
          RewriteEngine On
          RewriteCond %{REQUEST_FILENAME} !-f
         RewriteRule ^ index.php [L]
         </IfModule>
       </Directory>
</VirtualHost>

<VirtualHost *:80>
        ServerName www.luketimoth.me
        DocumentRoot /var/www/shattered-plays/public
        ServerAlias shattered.luketimoth.me

        <Directory /var/www/shattered-plays/public>
          PassengerEnabled on
        </Directory>
</VirtualHost>

As far as I am aware (with my beginner/rudimentary knowledge of Apache) this should do what I am expecting it to... but it's not. What am I doing wrong?

Output of httpd -S

httpd: apr_sockaddr_info_get() failed for LMTServe
httpd: Could not reliably determine the server's fully qualified domain name, using 127.0.0.1 for ServerName
[Sat May 17 16:42:02 2014] [warn] _default_ VirtualHost overlap on port 80, the first has precedence
VirtualHost configuration:
wildcard NameVirtualHosts and _default_ servers:
*:80                   www.luketimoth.me (/etc/httpd/conf/httpd.conf:1017)
*:80                   www.luketimoth.me (/etc/httpd/conf/httpd.conf:1031)
Syntax OK

EDIT:

I have changed the virtual host to use the desired subdomain as the ServerName, but I am still seeing the same problem.

<VirtualHost *:80>
        ServerName www.luketimoth.me
        DocumentRoot /var/www/luketimothy/public

       <Directory /var/www/luketimothy/public>
          <IfModule mod_rewrite.c>
          Options -MultiViews
          RewriteEngine On
          RewriteCond %{REQUEST_FILENAME} !-f
         RewriteRule ^ index.php [L]
         </IfModule>
       </Directory>
</VirtualHost>

<VirtualHost *:80>
        ServerName shattered.luketimoth.me
        DocumentRoot /var/www/shattered-plays/public

        <Directory /var/www/shattered-plays/public>
          PassengerEnabled on
        </Directory>
</VirtualHost>

httpd: apr_sockaddr_info_get() failed for LMTServe
httpd: Could not reliably determine the server's fully qualified domain name, using 127.0.0.1 for ServerName
[Sun May 18 10:15:24 2014] [warn] _default_ VirtualHost overlap on port 80, the first has precedence
VirtualHost configuration:
wildcard NameVirtualHosts and _default_ servers:
*:80                   www.luketimoth.me (/etc/httpd/conf/httpd.conf:1017)
*:80                   shattered.luketimoth.me (/etc/httpd/conf/httpd.conf:1031)
Syntax OK

There may well be a better solution to this problem, and I welcome those as well.

At the moment, I have an e-mail form on my website where people can send an e-mail directly to my gmail account through a PHP script. However, when I receive those e-mails, sometimes gmail flags it as possibly spam or phishing; presumably because the e-mail didn't come from a known server.

I was just wondering if there was anyway to tell gmail that if it comes from my server, it is safe. I thought that maybe I could add a unique custom header and tell gmail that that means it's a safe source?

I'm not very familiar with the deeper complexities of e-mail and e-mail servers, so any help would be much appreciated.