Tag Info

New answers tagged

1

To make shutdown work from outside the guest on CentOS 6, you need to install the acpid package and start the acpid service. yum install acpid chkconfig acpid on service acpid start


0

NGINX is most likely not properly installed, instead of compiling the code yourself, you can just use yum that will install it for you. It will be much easier and you will most likely not have these errors. To install via yum, I suggest you do the following: Destroy this droplet (back up important data) Create a new droplet and SSH in sudo yum install ...


0

As it turns out, Gmail was under the impression that the mailman emails were spam...


0

Your ServerName "reptool.dev:443" doesn't look right to me, it should not have the port in it. Try: ServerName reptool.dev


0

Actually, I just re-read this Oracle Blog post: https://blogs.oracle.com/wim/entry/fast_reboots_for_oracle_linux So, as long as you use the UEK kernel, you can do the kexec fast reboot on OEL5 and OEL6. This is not related to ksplice (which is about not rebooting in the first place!) and not needed a license or anything to use.


0

php-soap is in EPEL, so I wouldn't bother with compiling it. Install like this: yum install epel-release yum makecache yum install php-soap


0

Whay I ended up doing and worked for me: yum list installed | grep @rpmforge rpm --erase --nodeps <list obtained with first command> Then disabled rpmforge repos from the config files yum install <list obtained with first command>


2

If SELinux is in Enforcing mode, it won't let you do that... Try changing this boolean to true: setsebool -P httpd_enable_homedirs on Please note, I don't recommend disabling SELinux! There are many tools out there which can help you use SELinux to secure your system. Please see the CentOS documentation here: http://wiki.centos.org/HowTos/SELinux ...


1

Use yum history list to see recent yum transactions. Use yum history info <ID> to view the details of a transaction. Use yum history undo last to revert the most recent transaction, or yum history undo <ID> to revert a specific transaction.


-1

The files /etc/pki/tls/certs/localhost.crt and /etc/pki/tls/private/localhost.key are created by the postinstall script of the mod_ssl package. You can find the spec file here: https://git.centos.org/blob/rpms!!httpd.git/6f7de23710cf2fb30178692d962a51a8252842f6/SPECS!httpd.spec. Here is the script from CentOS 7: postinstall scriptlet (using /bin/sh): umask ...


0

You need to change your hosts_require_auth and hosts_require_tls options to the following: hosts_require_auth = <; $host_address hosts_require_tls = <; $host_address The problem will occur when using IPv6 to send. When sending with IPv6, $host_address will be an IPv6 address and will therefore contain colon characters (:). Colon is usually used ...


4

By default cloud-init expects to receive metadata from an Amazon EC2-compatible metadata service, such as that included with OpenStack and possibly other services. If you aren't running your VM under such a service, you have two options: Disable or uninstall cloud-init. This is the easiest option and if you aren't running in a cloud service then you ...


2

You can not attach two different security labels to the same directories. So, you had to select a single label and let the other process to read/write the files with that label. Problem is that doing that you are effectively granting the second process read/write access to the files owned by the first process, which is exactly what you don't want to do. ...


2

It would have been helpful to see the state of your iptables while you are experiencing the problem. fail2ban doesnt actually perform the blocking of connection requests, it only adds the rules to the iptables firewall when it detects a bannable action, as defined in your jail.conf.local. So you should first be checking the state of iptables sudo iptables ...


0

In case you want to know which packages are excluded due to priority protections you can use yum list updates -d3


0

You can use a GeoIP service to obtain the list of known China IP, and then forbid their connection with iptables. Have a look here GeoLite2 Free Downloadable Databases and download the CSV file. From the file GeoLite2-Country-Locations-en.csv you can find that China has a geoname id of 1814991. You then use that value to filter ...


-1

Well, get the IP range of all public IPs in China, add them to iptables with DROP. Stupid idea though, you should be using some DDOS-avoiding options (for apache for example: mod_evasive or mod_security) or even plain old limits in iptables like: iptables -A INPUT -p tcp --dport 80 -m state --state NEW -m limit --limit 50/minute --limit-burst 200 -j ...


4

If you're trying to mitigate DDOS attacks, then you need to have your ISP block the bad actors at their level. Once the DDOS traffic has hit your router/firewall/server, it's already eating up your bandwidth.


2

Replace -i eth0:1 with -d 1.1.1.1/32 I believe you have used ifconfig to create interface alias. Actually you can assign more than one IP for interface with iproute2 utility(ip addr add ..). It's better to use IP addresses in iptables rules.


0

To see what optical devices are available, you can run this as root: % wodim --devices As for "reading the values and apply it on to the application" you need to be more specific. What exactly are you trying to accomplish?


0

I've just stumbled on this question with a similar problem myself. I'd been trying to setup an LDAP server with selinux enabled and got a similar error. If you have selinux enabled, try it in permissive mode or disable it and try again. Not so good if you're trying for a secure system and ldap complains about selinux!


0

SSSD can be configured to retrieve user information from the Active Directory Global Catalog. That would require port 3268 https://fedorahosted.org/sssd/wiki/Configuring_sssd_with_ad_server If you will access any SAMBA shares, then dynamic ports will be necessary to verify access to folders before opening them. This TechNet document lists all potential ...


0

You don't mention which OS this is working with, so I'll keep it generic. It's possible that the mod_rewrite module isn't enabled and so the config is ignored due to the <ifModule mod_rewrite.c> You need to confirm that this module is actually loaded. Also you don't need two sets of <IfModule ...> This currently works for me: <IfModule ...


0

After trying so many methods I searched on net related iptables, I got one best solution, now everything is working like a charm. And applied below command with root access using putty. 1)First I tried this commands: iptables -A INPUT -p tcp --dport 2030 -j ACCEPT //press enter service iptables restart //press enter 2)then below commands: iptables -I ...


4

Most likely, you can't do this as many (most) providers will not delegate reverse DNS to their customers and offer some other way to edit reverse entries (if at all). When in doubt, check with your providers docs/support.


0

If you simply upgrade using rpm -U, the binaries will be upgraded, but the configuration files will remain the same. See http://stackoverflow.com/questions/14511334/rpm-upgrade-handling-of-config-files


2

If mysqld is running (check your process table), you should probably stop it via a kill command (dont use kill -9 as that will do an un-graceful stop of the service). When mysqld is no longer present in the process table, start the service normally.


1

In case someone has a similar problem, this is what I did. I hope it's the correct approach but I am not sure. The + in the permissions drwxrwxr-x+ indicates that there is an access control list in effect on the directory. I changed the permissions recursively for ACL as per below as well as included execution permission for the user. setfacl -Rm ...


-1

Enter top Then watch the httpd process. On the very left in the USER column it should say something like 'nobody' or 'www-userdata' or similar. Remember it. Then change to your home directory e.g. /home/myuser/public_html. Run cd /home/myuser/ Run chown -R nobody:nobody public_html (replace nobody:nobody if different user runs apache process) ...


0

This really depends on SSSD configuration, in particular auth_provider. auth_provider=ldap requires either port 389 (with TLS) or 636 (ldaps). auth_provider=krb5 requires port 88. ipa and AD providers require both actually, because even identity data is encrypted with GSSAPI, so you need port 88 to prime the ccache to do a GSSAPI LDAP bind, then port 389 to ...


9

You should use TCP ports 389 and/or 636. Port 636 is for LDAPS, which is LDAP over SSL. Encryption on port 389 is also possible using the STARTTLS mechanism, but in that case you should explicitly verify that encryption is being done. Microsoft's KB article says: Start TLS extended request LDAPS communication occurs over port TCP 636. LDAPS ...


3

As long as it LDAP auth only (and not AD/Kerberos etc.), 389 should be sufficient.


0

You can follow up my instruction. This works for me on Centos 7.0 & 7.1 and has not been tested on Centos 6.X. Step 1: Install pam_mount and cifs-utils. Step 2: Please only configure file /etc/pam.d/password-auth, just add 2 lines ( one is on top, and the one is at the end of the file) auth required pam_mount.so auth required ...


0

They are packaged by different maintainers and may conflict in the sense that they provide the same files etc, and possibly mysql rpm:s from official repos will not consider themselves upgrades of the mysql rpm:s installed from remi. I think dump the databases, uninstall all mysql rpm:s from remi repo, disable remi repo, yum clean all, install mysql from ...


1

Boot to just a command line by adding init=/bin/bash to the kernel arguments in the boot loader. Open a shell in a new VT by typing openvt -- /bin/bash Continue the original boot sequence by typing exec /sbin/init in the first bash shell. Once the boot sequence stall you can switch to the shell opened by openvt and start collecting information about the ...


0

Some pre-Westmere Intel CPU have a somewhat flawed TSC capability, maybe this is your case and the kernel is blacklisting the TSC as a timer source. Issue the following command dmesg | grep -i tsc and report here the output.


1

Well, I would start off with the naive approach like so: $ top |grep %pid On ubuntu 12.04, it outputs the line for the PID once per update of top: username@usrver / $ top |grep 2593 2593 username 20 0 1103m 170m 16m S 30 2.2 67:25.38 chrome 2593 username 20 0 1103m 170m 16m S 28 2.2 67:26.24 chrome 2593 ...


0

Now with the changed topic, this becomes a question for iptables. What I currently use (in overly simplified form) is: iptables -I INPUT -m tcp -p tcp --syn --dport 80 -m connlimit --connlimit-above 20 -j LOG --log-prefix "DoS Filter(connlimit): " iptables -I INPUT -m tcp -p tcp --syn --dport 80 -m connlimit --connlimit-above 20 -j DROP obviously you may ...


3

Currently the proc filesystem is not "container aware" in mount namespaces, so tools basing their logic on this will get host-related values instead of container-related values. But a work is in progess, it's called lxc-fs and few releases are available here. This is a user-space workaround that will make possible a bind mount over /proc to get things ...


1

Connection timeouts are most often caused by packet filter or NAT device timing out your TCP connection due to inactivity. There is no one golden rule regarding ssh settings. Much depends on your systems and connections. Generally you should set keep-alive interval value to less than it usually takes for your connection to time out (be kicked off from the ...


4

There appears to be no way around it. LXC uses cgroups to do its RAM-limiting, and non-virtualization aware tools read stats like /proc/meminfo which is not contained within LXC and will output the overall RAM in the system. You can see this behavior with free or top as well when run inside the container. Source: ...


0

how about using CentOS Software collections? Those are specifically designed for parallel version installs. I would also suggest getting familiar with the process of building those so that you can create your own if needed. Use Copr for build infrastructure as needed.


0

You can build ImageMagick extension for latest version of PHP using below steps. Download source form https://pecl.php.net/package/imagick/download Unpack archive and go to unpacked folder. #cd imagemagick-x.x #path_to_php_home_dir/bin/phpize #./configure --with-imagick=/opt/imagemagick #make #make install imagick.so will be build in module ...



Top 50 recent answers are included