New answers tagged

0

Somewhat of a late response, but just faced the same problem with CentOS 7 minimal install. Solution was to run: grub2-mkconfig -o /boot/grub2/grub.cfg after upgrading the kernel. Note that does not set the new kernel as the default in the boot menu.


0

Turns out it was a firewall issue after all.. OVH has a really buggy web-based firewall which was causing the problems. The staff at OVH told me that my firewall was set up 100% fine, because otherwise it would not be possible to resolve using 213.186.33.99. This was incorrect, the firewall is bypassed for this specific IP because it is apparently hosted ...


1

More than likely you're pointed to 127.0.0.1 (localhost) because your distro uses a caching name service daemon. Think of it like a caching DNS server that runs on your own computer. The use of such a daemon is of debatable effectiveness. It does have some benefits in speeding up some types of Internet usage by cutting down on the number of DNS requests ...


3

127.0.0.1 is your localhost loopback interface, meaning you're reaching your own server. For various reasons, your server is using that first to lookup DNS and since your server doesn't know how to answer a DNS query, you have to wait until the query times out and moves on to the 2nd nameserver. Just use the real DNS nameserver as the only entry: ...


-1

Reverse the name server order and it will speed up. nameserver 213.186.33.99 nameserver 127.0.0.1


0

Many thanks to @larsks for getting started. I found that I had trouble with the script completing before it was needed so my issue file was always blank. So I added a 1 second pause and looped. I'm not really a bash scripter, so if there is a better way to do this please let me know. I added a COUNT to make sure it didn't loop forever. #!/bin/sh COUNT=1 ...


2

This is probably due to OS support or a resource issue. EL7 was not intended for use with vSphere 4. The VMware support matrix reinforces this. I see you're using open-vm-tools, but it looks like you may have a deeper issue. See: https://access.redhat.com/solutions/21849 and: https://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=...


0

I realized it is a "hairpin DNS" issue. Best solution I guess to configure NAT to not shortcut request to my own external IP to local IP... but still my provider not did it and I temporary fixed it by nano /etc/hosts added a line y.y.y.y myserver.com where y.y.y.y is my LOCAL address.


1

Not quite the answer to your current problem, but it will render the whole issue moot since you're trying to trigger the script from the same host running your site anyway: When you need to run PHP scripts from cron you're generally better of running them from the PHP command line interface rather than making a web request to the script. Depending on the ...


1

Take a good look at the actual running iptables configuration ipatables -L -vn and ensure that nothing is blocking the source IP of the system running he curl/ping commands - take appropriate action. If there is nothing there then something upstream, between the systems may be blocking. As ping is being blocked I guess you could try mtr mtr -4 example....


0

i opened a ticket in github and the answer is: Pls remove the networks that are conflicting using docker network rm (or) stop the deamon, remove the /var/lib/docker/network directory and restart docker daemon.


2

An approach is to serve static files in the front-end and only proxy if they do not exist. You can deploy your error pages and resources on this server. Avoid directory conflicts between the front and back ends. For example: root /usr/share/nginx/html; location / { try_files $uri @proxy; } location @proxy { proxy_pass http://api; } See this ...


2

You're right - nginx can't distinguish a request for the error page assets from normal requests. You should either change the assets to use a unique path that you can do location matching on (maybe something like /500/logo.png), or use an external service to host the assets (S3 or the like), and load it from there. For the location match, it would ...


4

Changing the webserver port also requires changing the SELinux policy for the webserver which is well documented : Run the semanage port -a -t http_port_t -p tcp 82 command as the root user to add the port to SELinux policy configuration. Run the semanage port -l | grep -w http_port_t command as the root user to confirm the port is added... ...


2

No, this is not possible. Haproxy does not understand the FTP protocol, and therefore can't do things like re-write the FTP commands and responses that include the true IP address of the FTP server. If that's a requirement, then you'll have to find a different proxy solution.


1

Apache was already installed with Cpanel on CentOS 7.2. Create folder in /home and set ownership to root:nobody to allow apache to access it. That should do the trick.


0

On CentOS 7.2 /var/www/html is automatically created when installing Apache httpd and is functionally similar to Ubuntu's directory: [root@t-wiki ~]# rpm -ql httpd | grep /var/www/html /var/www/html [root@t-wiki ~]# ls -l /var/www/html total 0 [root@t-wiki ~]# rpm -q httpd httpd-2.4.6-40.el7_2.1.x86_64


-2

If you server is also the outgoing SMTP server for PHP, it maybe trying to deliver messages local-to-local. From what you've said, it has an valid domain registered to it. Basically, your root@domain.com might be working as a domain catch-all address. (Educated guess based on experiences with hMailServer on Win Server 2012)


0

Answering myself: I just found out that the PDF will be saved in the home directory. This means: /home/Usuariotest


3

You don't. But I have very little knowledge about server. and creating a name server are not compatible. Getting DNS right is quite complicated and really a waste of time as you get a large number of providers that do that for you for a few dollars a year.


3

This trick helped a colleague who did something similar with another critical package. If your KVM access allows to paste from your clipboard then you could copy over libssl.so.10 & libcrypto.so.10 from a known working installation. To prevent your terminal getting confused, Base64-encode the files first. Something like this on a known good system: $ ...


3

If you have no way of getting a physical media attached to the server, then you need to get the server wiped and reinstalled. Since you managed to corrupt all cryptography libraries in the OS, you need to replace them with correct versions from physical media. However, the recovery method above requires quite extensive knowledge on how your Linux ...


0

That error is addressed in https://openvpn.net/index.php/open-source/faq/79-client/253-tls-error-tls-key-negotiation-failed-to-occur-within-60-seconds-check-your-network-connectivity.html and basically indicates UDP/TCP 1194 is not able to fully reach through your router to the OpenVPN service. I see step 5 in the directions you followed has you stop ...


22

Two ways: Use limits.conf to assign the maximum allotted memory per process for that user Create a cgroup for that user in order to limit their total memory usage More details here: http://unix.stackexchange.com/questions/34334/how-to-create-a-user-with-limited-ram-usage


4

If you have home directories on a separate partition you can mount the partition with noexec. Although I'd also agree to just ban the user.


10

You are trying to solve a human problem via technical means. This person is knowingly violating policy. The appropriate response is to (as @ceejayoz wisely mentioned above) get rid of that user, or at the very least, remove access to the system in question and any others like it.


2

rotate 0 from the man page rotate count Log files are rotated count times before being removed or mailed to the address specified in a mail directive. If count is 0, old versions are removed rather than rotated. You probably want rotate 9 As to why you are seeing all the other files, my guess is that they are old (check the date/time) and ...


0

For internet access to work behind a NAT instance, your default route in your private subnets needs to be the internal ENI of your NAT instance - see docs. Also, your NAT EC2 instance needs to have the Source/Destination Check disabled - see here. I would also make sure you have followed and understood the Amazon documentation on using a NAT EC2 instance. ...


0

This kind of looks like something important got removed with the package installation. the errors relate to policy kit. Can you confirm that polkit and dbus are both still installed? rpm -qi dbusand rpm -qi polkit logs may also contain information but seems here a policy is preventing the start of these services due to dbus not responding...because dbus is ...


0

When you install docker it adds a bridge to connect to the outside world and an IP class 172.x.0.0. You can change that by following the document here https://docs.docker.com/v1.10/engine/userguide/networking/default_network/custom-docker0/ . If it disconnects you, look for the docker file /etc/default/docker or /etc/sysconfig/docker.


0

where is your supervisord.conf stored ?. Can it be that your supervisord was upgraded/reinstall and placed a new conf file somewhere else ? and you're referring at start-up an older version of the conf file ?.


0

I've managed to solve the mystery by deleting the server context from nginx.conf. It comes after the include directive, and I guess it overshadowed the server and upstream blocks from myapp.conf, although in the output of nginx -T, the latter appears last.


1

nginx -T will perform syntax check and dump your final configuration file to stdout. https://nginx.org/en/docs/switches.html -T — same as -t, but additionally dump configuration files to standard output (1.9.2). Also you could check error log /var/log/nginx/error.log for some runtime errors(can't bind to address for example).


1

The solution was derived from a post by Radek Simko: wget ftp://xxmlsoft.org/libxml2/libxml2-2.7.8.tar.gz tar -xvf libxml2-2.7.8.tar.gz cd libxml2-2.7.8 ./configure make sudo make install It was a simple as compiling from source for an install. I could then: install.packages("igraph") In R successfully.


0

I managed to overcome this issue by downloading the correct devel repo and having the repo for the dependencies also enabled , the devel repo has the following content : [root@localhost ~]# cat /etc/yum.repos.d/puppetlabs-devel.repo [puppetlabs-devel] name=Puppet Labs Devel 6 - $basearch baseurl=http://yum.puppetlabs.com/el/7/products/$basearch gpgkey=file:...


1

Sounds like you've suffered some sort of soft disk failure that has triggered a check/rebuild of the disks and that failure corrupted part of the dovecot installation. I would also: If possible, check the S.M.A.R.T. performance/health counters for your disks, there might be a catastrophic hardware failure imminent Verify the files of other packages using ...


0

Reinstalling dovecot packet fixed the problem: # yum reinstall dovecot # service dovecot restart Update: smart doesn't report any error.


0

I would recommend verifying the package that owns that file against the RPM database. You can do this with rpm -Vf <file>. This should identify any changes that are not "stock". More information can be on RPM verification can be found here, but here's an excerpt: Now you can let RPM do at least part of that sleuthing for you. Simply direct RPM to ...


0

It it literally trying to email root@ with no domain which is obviously invalid (thus the error domain "" is syntactically invalid). There are several steps exim4 should be taking in order to get from your invalid email address to a valid one. First, exim4 checks to see if you should be allowed to submit an unqualified email. From the documentation, ...


0

Have you made any changes to /etc/rsyslog.conf? This conf files contains the settings for your logs, you should definitely check that and if you want, post it's contents here and we can help you review it.


4

You can use IPTables for it, just set allow rules for your whitelisted IPs and reject all others.


0

Problem For me the pid file name was different in those two files: /usr/lib/systemd/system/nginx.service pid /var/run/nginx.pid; /etc/nginx/nginx.conf PIDFile=/run/nginx.pid Those two need to match. Fix: So I adjusted it in /usr/lib/systemd/system/nginx.service and then did: systemctl daemon-reload systemctl start nginx Then it came up ...


0

Easiest way to do it, to install it from Remi repo. Here wizard, where you can see commands to configure your system. After configuration, just run yum install php-gmp.


0

The problem is you're using OpenVZ. OpenVZ is running a 2.6 kernel which does not have the capabilities that the firewalld daemon relies upon and all of the systemd changes are backported into sysvinit for OpenVZ.


0

What's the problem? All seems correct. Remove Nginx. Then restart Apache and then restart Varnish.


0

The error you've provided doesn't match the repo configuration you've pasted - the error suggests you've got some other definition of puppetlabs-devel that contains some ERB (<% etc). I'd suggest searching /etc/yum.conf and /etc/yum.repos.d/ more thoroughly and delete any other instances of the repository (grep -r puppetlabs-devel /etc/yum*). You should ...


0

I figured out my problem. In the midst of troubleshooting, I had compiled "libmemcached" which appears to be outdated not "memcached" - it appears to be running now!


1

What am I missing? That memcached needs to be in your path.


2

With nslcd -d I saw that the address was already taken by the host system. I fixed it by mounting the socket when executing docker run with -v /var/run/nslcd/socket:/var/run/nslcd/socket


0

You can use yum update-minimal update-minimal Works like upgrade, but goes to the 'newest' package match which fixes a problem that affects your system



Top 50 recent answers are included