Is there any way I can change the .NET Runtime of a website without full server app domain recycle?
It appears when changing the .NET version in the configuration properties, all of our app domains appear to be restarting, but stop and must manually started. This is annoying. Appcmd.exe would be a great solution, but this is only available on IIS 7 :(
How do I check what version of syslinux I have?
nmap -eeth1 [hostname] -p80
Running that command as root uses interface one and fails as expected (interface one is on a subnet blocked by the host.) Using interface 0 succeeds as it should.
However, if I switch to user zabbix then the above command succeeds no matter which valid interface I specify. If I specify a nonexistent interface it fails as expected.
Any idea why the output is different between root and the zabbix user?
My current settings:
query_cache_size=20M
query_cache_type=1
query_cache_limit=1M
Most of the queries return the exact same set of results (about 15 rows) and occur every few seconds.
Can anyone tell me how to interpret this graph for cache tuning?
Basically I am just wondering if I can drag and drop the libphp5.so file to another server without PHP and have the functionality of PHP and all the extensions statically compiled into it. (Compiled from source with a bunch of extensions)
I tried using the ld tool on the so file (on my server) but it just spit out a bunch of undefined references. Which are irrelevant since its a working install.
Edit:
I realize now that I was using ld instead of ldd.
Now I get a list of the dependencies. I am guessing I would need all these files correct? So my question is if I recursively get all the shared object files and put them in their same locations on a destination server (probably via a custom RPM) would PHP function a new server without a formal install of PHP?
Could I be missing any hidden dependencies that are not found by ldd?
With Redhat and CentOS (especially bugs and patches) I often see the terminology upstream and downstream. What do these terms mean in this context?
Example: Red Hat would like to thank the X.Org security team for reporting this issue. Upstream acknowledges Rafal Wojtczuk as the original reporter.
What is it with server hardware that actually makes it more suited for server hosting vs. say a collection of cheap PCs or a good PC?
I am investigating methods of monitoring particular stats with printers. I am not very knowledgeable yet, and was just curious what .mib files are used for and why there are many of them for a printer?
After some google research I have figured out inode usage probably relates to the number of directorys/files.
I just want to capture the output of a time command i.e:
X=$(time ls)
or
$(time ls) | grep real
The time function spits it to the console though. How do I do this?
There is a very similar question to this, but there is an extra detail I have in my situation.
Basically I am using a ControlMaster to speed up SSH connections. So I am using:
CurrentUser=$(whoami)
session=$(ps -el | grep $(id -u) |sed -e 's/sshd//g'| grep ssh)
if [ "$session" = "" ]
then
ssh -TMNf [email protected] &
fi
After this code I have another command. The purpose of the ssh command is solely to run in the background in order to keep an open socket for faster connections (if no calls occur for a short time then it times out and closes due to inactivity.)
The problem is this script doesn't close and return to the caller. It becomes a < defunct > process, and doesn't return until I manually kill the created ssh session. How can I let the script return to the caller and have the created ssh run in the background?
Edit: I want the created ssh session to persist after the script ends to keep a socket open for faster connections with an rsync command after this. With an open socket I go from 3-4 second transfer times to milliseconds
On our Repository server adding a ControlMaster to the ssd/config file (creates socket?) resulted in a 50x transfer speed increase from roughly 2.5 seconds averaged to 0.05 seconds average. The transfer speed is between the repository system and our web servers. Now I would like to apply the same performance gain to the windows -> repository connection. We currently use Tortoise to connect via SSH, and I read on wikipedia that Putty does not natively support session multiplexing. I am aware there are other SSH tools with the feature that are also made for windows, but I was wondering if using these if I would run into issues with compatibility with Mercurial using the SSH connection as right now I believe it uses Plink to handle the SSH tunneling.
I am looking for any options for enabling session multiplexing on windows that are compatible with Mercurial.
I often have problems with SELinux, most of which I have resolved eventually. There are still some situations where doing custom policy would be very elegant and more proper. I'd like to get the skillset of being able to fully do with SELinux what I want. I would ideally like to eventually be capable of implementing a strict policy myself.
I am looking for advice on where to start or how to get into SELinux. It seems like Dan Walsh and Stephen Smalley are the only ones on the internet who seem to understand everything (They are the names I always see on half to fully resolved SELinux issues).
I am not looking for simply a link and "Go here and read" I am looking for clear introductory material with a context of just starting with SELinux - not a technical manual. If you understand it how did you get started?
If this should be community wiki, let me know not sure if it should be or not as it's semi-subjective.
I don't want references to automated tools, I want custom policy. I want to get close and dirty with it, but without having to climb a cliff as the learning curve.
This might be better for super user, but I figure as a tool of the trade it might be a better topic for here. I find often when editing in vim that I exit out, run some script that I am testing, and then go back into vim (yes I realize I can use :!command to run things from vim) I was just wondering if there is a way to hook something to run when the file is saved (:w)?
SELinux is not allowing me to change the context of a symlink. I get no errors or anything. It just doesn't change the symlink, which I am guessing means it is resolving the symlink and changing that instead? coderoot is a symlink:
chcon -t httpd_sys_content_t /var/coderoot
This leaves the context of the symlink the same. When my application references the symlink SELinux jumps in and says no. With SElinux off it works fine. How can I fix this?
I have been having a lot of fun coding with what puppet I have had experience with so far, and so far it looks like it can do most of what I need it to do.
One thing I am not finding in the documentation (with some searching) is how to (if it is possible) have pre-built configuration files to use on the servers. For example:
I use puppet to install httpd. Now we have apache on the client. Well now the httpd.conf needs some customization and specific settings. I have the httpd.conf I need to use on the puppet master. How do I get it from the puppet master to the client?
http://www.samba.org/samba/security/CVE-2010-2063.html
How can I check to see if the Redhat (CentOS) repositories have backported a fix?
I have looked at the official sites documentation, and it is a little tough to find starting points.
Is there a free comprehensive guide that is easy to understand for someone that has never dealt with mod_security?
In the configuration I have setup I wish to allow samba and apache to access /var/www I am able to set a context to allow samba access, but then httpd doesn't have access. Using setenforce to 0 eliminates issues so I know that it is SELinux.
In addition: How can I view the context of a folder, and can a folder have multiple contexts?
(CentOS)
We are using PHP,MySQL,SVN, and Apache. I want to know if I assume the local subnet trusted, and allow all ports on our subnet using iptables, then allow only port 80 as open to "anyone". If it is "good security" taking that the subnet is trusted as a premise. This is also assuming good PHP code etc, and that apache is configured correctly.
This is using CentOS 5.