New answers tagged

2

You should check if yum is configured to check gpg signatures. You're looking for something like this in /etc/yum.repos.d/epel.repo: [epel] [...] gpgcheck=1 If that's there, then as Alex notes, you wouldn't be able to use the repo if the signatures weren't matching the keys. That is also Red Hat's formal advice on the subject (see bottom of linked ...


0

If it wasn't working, you wouldn't be able to use epel repo, as simple as that.


0

It depends... Do you know anything about the hardware and its BIOS settings? That's where I'd check first. Outside of that, you can modify the system's tuned-adm performance profile. See the chart here. (see the CPU governor)


2

If you've arrived here, chances are you have two services where one depends upon the other but, because they're starting in the wrong order, the one with the dependency is failing to start. Suggestions about editing the symlinks are informative, in terms of illustrating how the startup sequence runs, and would work alright until someone did a "chkconfig on" ...


0

I presume that the machines have different version of the same OS or even different OS. You could manage different version of Python using for example virtualenv or virtualenvwrapper. After that you install the packages you want in your /env installation/ more info about virtualenv


3

Something like this should work - this is bare bones only. This is hardly a professional question though, this is an absolute basic of Nginx that you should be able to work out yourself with a simple tutorial. server { server_name example.com; listen 8085; location /site1 { proxy_pass http://example.com:8080; } location /site1/task { ...


1

Yes, this is an upstream Samba bug, which was introduced to RHEL users in the latest Samba package updates. Red Hat is aware of the problem, and has a candidate patch to fix it, but as of today (Apr 27) has not yet released an update with the patch. See https://bugzilla.redhat.com/show_bug.cgi?id=1326918 and ...


0

LVM should only be used on a single system, if two nodes think they have exclusive access to an LVM volume corruption will occur. HA-LVM allows for multiple nodes to read from a shared volume but assumes only one node can/should write - this is analogous to how Windows MSDTC clustering worked/works (not that up to speed with in today sorry) where a single ...


0

Red Hat has an extremely accurate documentation of their products. On your case at hand, please refer to https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/7/html/Linux_Domain_Identity_Authentication_and_Policy_Guide/prerequisites.html#dns-reqs If you are still experiencing troubles, pleas specify exactly what you have tried and what the ...


0

Hej all, I have solved it on my server (RHEL 7.2) by editing '/usr/lib/systemd/system/postgresql-9.5.service' and adding 'After=remote-fs.target' Hope this helps. Best, Tim.


1

I faced the same problem: AD authentication without winbind was broken for me. The reason why I stopped winbind originally was because it could not find the local groups (although it seems to find the correct user mapping). Here is a log except with winbind activated: [2016/04/22 16:15:20.654780, 5] Auth/token_util.c:527(debug_unix_user_token) UNIX ...


0

This is an old question that may need a fresh answer. Learn about spacecmd -- it'll save your brain and mouse-clickety hand some needless pain. AND, it can be scripted, even if it's a bit wordy. for CHNL in `spacecmd -q -- softwarechannel_list`; do spacecmd -q -- \ softwarechannel_listchildchannels ${CHNL} \ | xargs -n1 \ spacecmd -q -y -- \ ...


1

You should use Amazon SES. Maintaining an SMTP server on EC2 is expensive and not worth the effort. However, do keep in mind that Amazon encourages proper email behaviour so you should keep track of your bounces and remove such invalid email addresses. You might get your sending quota reduced ot even blocked if there are bounces. They have SES best ...


2

The IUS project provides RPMs for newer versions of select software (such as PHP, MySQL, and Python) for RHEL and CentOS. After adding the repo, you can install PHP 5.5 FPM by running this command: yum install php55u-fpm Alternatively if you want to use mod_php you can run this command: yum install php55u This packages conflict with and provide their ...


0

check max open files set for the user. You may need to increase that. use pidstat for the rpocessid (pidstat -h -r -u -p $javapid | grep -v "Command") and no of processes running (ps aux | grep 'java ' | grep "application name" | grep -v grep | wc -l). MUP (Max user process) set for the user is important.


0

So, this may not be the right place to put this, but I do not have the reputation points to comment. I am also having this issue with Scientific Linux 6.7 and upgraded last Wednesday across our systems. I found that the problem lies in the netbios name resolution (nmb). If the user specifies the fully qualified domain name of the active directory, they are ...


4

The ABI between CentOS 5 and 6 is not guaranteed to be identical. This means that your programs may work, but this is not sure at all. Moreover, any packages/libs upgrade can stop your application from work. In other words, I strongly suggest you to do not go down this road.


1

The infomix locale need to be set in crontab INFORMIXDIR=/home/informix CLIENT_LOCALE=es_es.8859-1 DB_LOCALE=es_es.8859-1 0 15 * * * /home/user/my-script.sh The locale es_es.8859-1 is in my case, you need to check what is your locale.


0

Edit the kernel booting options, change ro init=/sysroot/bin/sh to rw init=/sysroot/bin/sh and boot CentOS. When logged in, follow these commands: chroot /sysroot passwd root touch /.autorelabel reboot You now should be able to loggin as root user. Source


0

this feels like an selinux issue with the files related to passwd. If something has changed the passwd files when selinux was not active it can have the wrong labels and selinux will deny access to the file. (You can probably confirm if selinux is on by examining the kickstart file) Follow the instructions from here to boot into rescue mode. After you gain ...


1

you are in single user mode so you don't specify users make sure your root filesystem is mounted in read/write mount -o remount,rw / then passwd


1

No answer gave proper advice on how to update the SELinux rule as well as making the settings persist upon a server reboot, to do so do the following: # setsebool -P httpd_can_network_connect on # setsebool -P httpd_can_network_connect=true This was done on RHEL7.


1

This is a difficult question for us to answer because we don't know what your program does. Bottom line though is that java.lang.OutOfMemoryError is often caused by either spikes in traffic on your app or a memory leak. You should play around with the two java flags that specify the memory allocation pools for the Java Virtual Machine:-Xmx(maximum memory ...


1

You're specifying -Xmx128m when instantiating the JVM. Try increasing that value.


0

Seems like the following could be your issue here "When loaded together with mod_fastcgi, mod_auth must be loaded before mod_fastcgi. Or else users will experience long delays when login in and sysadmins will probably not find out the source of the problem due to the lack of meaningful error messages." From the Lighttpd documentation ...


1

you don't have to. You can use your existing dns and ntp server infrastructure but you will need to manage the ipa records on your on your own instead of having ipa take care of it. In fact it does not really matter what servers run the dns service as long as the clients get their answers from you the infrastructure in place. You can, for instance, delegate ...


0

In my experience this is usually the result of time drift on either the domain controller, or in your case with only one client having an issue, the connecting client machine. Because Kerberos includes time related parameters in both the authentication server request and the authentication server reply (AS_Req and AS_rep) a wide discrepancy will cause the ...


-1

Just use ">>" which means append instead of ">" which means create from your scripts that write to this file. I had the exact same issue and I fixed it using append in my script. SomeScript.sh >> output.txt Hope that is clearer.


0

You could try with this: Real machine Ensure you have net.ifnames=1 biosdevname=1 in grub.cfg file. Change /etc/sysconfig/network-script/ifcfg-enX to /etc/sysconfig/network-script/ifcfg-dummyX Ensure that the option DEVICE=dummyX in file /etc/sysconfig/network-script/ifcfg-dummyX (the option is very importance) Vmware machine: I do not test but maybe ...


1

DNS and NTP being synced between your IPA server and clients is critical for kerberos tickets to be issued and considered valid. That is why it is recommended by RedHat to set up your IPA server as an NTP server. DNS is more optional; the critical part of DNS is that forward and reverse lookups succeed and match. The RedHat published guide on IdM (built on ...


2

Some DHCP server software operate differently when it comes to reservations: some require that all reserved addresses be outside the dynamic range but be within the subnet. some require that all reserved addresses be within the dynamic range, and it then knows not to use these when allocating other hosts dynamically. some allow a reserved address to be ...


2

Static IPs are normally outside the pool range as you do not want static IPs assigned to other hosts. As long as the static IPs are in an IP range local (or relayed) to the DCHP server they can be served. The leases file records which addresses from the pool have been assigned to a host and when that assignment expires. This is used to ensure that ...


2

Yes, you can specify conditions on both (source) IP and (source) port in a single rule, both of which conditions must be satisfied in order to prevent communication, eg iptables -I INPUT -s 202.54.20.22 -p tcp --sport 5000 -j DROP Note that you must specify a protocol in order to use sport or dport, so if you wanted to block both TCP and UDP, two rules ...


0

If you want to just get the version numbers the following is about as short and simple as I can get it. Tested on rhel 6.7, rhel 7.2, debian 8.3 and ubuntu 14.04: lsb_release -s -r | cut -d '.' -f 1 For a practical example, say you want to test for the distribution major and minor version and do things based on that: #!/bin/bash major=$(lsb_release -s ...


0

No, this is not directly possible, as the client know nothing about the provisioned storage. You can infer some information using specific client-side benchmarks, though.


0

Try CloneZilla (http://clonezilla.org), it can clone partitions to smaller destinations.


1

You can use GParted to resize your RHEL5's partitions then use 'Volume Based Cloning' in Vmware's converter. This should work unless you have reason to keep your physical RHEL completely as is...


2

Back when I first started as a Linux admin 8 years ago I use to use a popular third party repository to upgrade my LAMP stack. It was ran by a single individual.One of the primary reasons was developers pressuring me for a newer version of PHP than what came with RHEL 5. It ended up biting me. The person abandoned the repositories so I was no longer ...


3

The question is not if we trust Andy, it is if you trust Andy. I'm not familiar with the repository but the donation button suggests a personal effort. Feel free to contribute if it has value to you. Packages look to be GnuPG signed, so it is possible to verify with some certainty the packages are authentic. You can also check if he is on the web of trust. ...


1

I've not run Cloud Linux, but this tutorial from Digital Ocean shows how to install it from source: visit the download page and choose the latest release tarball (1.7a at the time of this writing). After downloading the tarball, untar it into your current working directory: # tar zxvf /path/to/haveged-x.x.tar.gz Now you compile and install: # cd ...


0

Normally /etc/resolv.conf file is a configuration file for DNS related option (nameserver, search domain, ecc). In a Red-Hat based distribution I will suggest you to create two files: /etc/sysconfic/network-script/route-eth0 /etc/sysconfic/network-script/route-eth0:0 (I have not ever seen this) In each of one you could defined all the ...



Top 50 recent answers are included