Tag: SysAdmin

The post How to Check for Available Domains appeared first on Heart Internet.

The rules around SSL certificates are changing across the whole internet. The good news is that for most customers, very little will change on your side. This is an industry-wide... The post SSL Certificates are changing. Here’s what you need to know. appeared first on Heart Internet.

The post Hosting VPS Linux vs Windows VPS appeared first on Heart Internet.

The post Domain Name Transfer Checklist: Everything You Need to Know appeared first on Heart Internet.

We’re incredibly proud to celebrate our Site Reliability Engineering team, who have won the Gapstars Innovation Award for their outstanding work improving platform stability, security, and visibility across our shared... The post Heart Internet Win Gapstars Innovation Award 2026 appeared first on Heart Internet.

If you want to improve your website you probably need to do A/B testing, otherwise known as split testing. Instead of guessing, A/B testing allows you to experiment more scientifically.... The post A/B Testing Explained: A Practical Guide To Better Results | Part 1 appeared first on Heart Internet.

Account security matters, and switching on two-factor authentication (2FA) is a quick win. 2FA adds a second check during the sign-in process, so even if someone compromises your password, they still can’t get in.  To enable 2FA:  Step 1: Open your... The post How to enable two-factor authentication (2FA) on your Heart Internet account appeared first on Heart Internet.

Get Your Name Right – The Internet Never Forgets Choosing a domain name might sound simple – until you realise it’s the online equivalent of naming your child. No pressure.... The post How to Choose the Perfect Domain Name for Your Business appeared first on Heart Internet.

Discover what a VPS server is, how VPS hosting works, and why it’s ideal for small businesses. Learn the benefits and explore VPS plans with Heart Internet. The post What is a VPS? And is it Time You Got One? appeared first on Heart Internet.

💚 Hosting that works hard, treads lightly.   Big news: Heart Internet is now officially listed with the Green Web Foundation. That means our hosting services are recognised as being... The post We’re Now Certified by the Green Web Foundation appeared first on Heart Internet.

Looking through the docs its not clear to me if this is possible. I have previously use URL masked serverless backends to route to my cloud run services. This is easy to configure on the GCP load balancer and its a one time config and all future cloud run services that are deployed are automatically accessible via https://my-lb.mydom.com/service-abc. The problem is GRPC does not support path based routing. So I can't send GPRC requests to my-lb.mydom.com/service-abc:443. I don't want to use host based routing and I want to avoid Cloud Service Mesh or Traffic Director. Is there no way to support this with vanilla GCE load balancing? If I have to use Cloud Service Mesh and/or Traffic Director does it work with URL masked backends so it automatically routes by cloud run service name?

How to configure Proxmox host, which has Nvidia L40s GPU(1) for passthrough to a VM running on same Proxmox host. I have HPE Proloiant DL380 with Intel Xeon Processors. 1 Nvidia L40s GPU card. I want to acheive Pcie passthrough to a VM running on Proxmox.

I'm developing a daemon that runs under control of systemd, and that has a feature to reload code that's inbetween "systemctl reload" and "systemctl restart". Unlike "systemctl reload" that sends a signal to re-read just the configuration, it also reloads the code, but unlike "systemctl restart" it's not a true hard restart that forgets all the state, but instead the state is written to a file and a signal causes the daemon to replace its code with execve() and then it reads the old state from the file. Is there any feature in systemd that would allow me to add some custom command to systemctl that would be inbetween reload and restart, just for this daemon? Technically the feature would be implemented by some signal such as SIGUSR1 or SIGUSR2. I'm expecting that there could be cases where a hard restart will be done instead of the lighterweight "dump state + execve + reload state", so it would be useful to have

I've got a storage server with some NVMe SSDs for system and metadata cache and 24 3.5" HDDs for storing historic data. Since this data is very infrequently accessed (at least once per day for synchronization, often that is the only access at all for several days), it would be good to spin down the drives during the remaining time to conserve power. The system is running Debian (actually Proxmox, but no VMs there) and ZFS. The storage pool is made up of 3x 8-way raidz1, with a mirrored special device on NVMes for metadata storage. The disks are Seagate Exos SAS drives. My questions: Will the drive lifetime decrease (or increase) if they are only active maybe 1/4 of the time, but with at least one spin-down/spin-up every day? Can or do I need to tell ZFS about this behavior, so that it isn't confused by very long initial access times when a drive needs to spin up first? Does ZFS even let a drive go to sleep, or does

What are the use cases for libnss-extrausers? I've seen a couple of examples but the details are lacking. If you use libnss-extrausers, don't you have to change all user and group IDs to keep from colliding with the corresponding /etc files? If this is done then effectively only files with "world" access in directories with "world" access are available. What about home directories, do you use a common directory such as /tmp or what do you do? This is why I'm asking about the use cases to better understand the benefit.

I have several java webapps running on 2 tomcat servers--with various JDK (17 and 25) and Tomcat versions (9 and 11), in a test/development environment--where the server is configured to request a client certificate. When the browser prompts me, I select my CAC card certificate on the browser popup and then am prompted to enter my pin. The prompts for certificate selection and pin entry have always occurred immediately after navigating to the url (so there is no post-handshake authentication). Unfortunately, this stopped working for me; one day it was working and another day it totally stopped. I am 100% certain that no server nor personal configurations changed since the time it was working; no server, connector, JDK, or any other settings were manually changed. And even more strange, it is very inconsistent between users: it still works 100% for one team member, it works only on Firefox for another team member, a third memeber can only get it to sometimes work in Chrome's p

I have a fleet of Dell OptiPlex 3000 machines that were purchased due to one reason only: legal license for Windows 10 IoT Enterprise LTSC. I am getting very strange set of permissions on a clean re-install on certain folders that relate to the shared Start Menu: C:\Users\All Users\Start Menu\Programs S-1-5-21-3671523672-3566060235-3176437112-1000:(I)(OI)(CI)(DE,DC) DESKTOP-UROR7BK\admin:(I)(OI)(CI)(DE,DC) NT AUTHORITY\SYSTEM:(I)(OI)(CI)(F) BUILTIN\Administrators:(I)(OI)(CI)(F) BUILTIN\Users:(I)(OI)(CI)(RX) Everyone:(I)(OI)(CI)(RX) C:\Users\All Users\Start Menu Everyone:(DENY)(S,RD) Everyone:(RX) NT AUTHORITY\SYSTEM:(F) BUILTIN\Administrators:(F) C:\Users\All Users NT AUTHORITY\SYSTEM:(OI)(CI)(F) BUILTIN\Administrators:(OI)(CI)(F) CREATOR OWNER:(OI)(CI)(IO)(F) BUILTIN\Users:(OI)(CI)(RX) BUILTIN\Users:(CI)(WD,AD,WEA,WA) C:\Users NT AUTHORITY\SYSTEM:(OI)(CI)(F) BUILTIN\Administrators:(OI)(CI)(F) BUILTIN\Users:(RX) BUILTIN\Users:(OI)(CI)(IO)(

I have a windows NFS setup with AD and on mounting it in linux vm. I’m able to create and edit the files but not able to delete the files. The user seems to be correctly mapped but delete fails. I have even given full control to the user but it still doesn’t work. Can someone help me understand the possible causes for this? The delete works correctly on the windows machine.

We have a pair of Windows Server 2019 servers running Microsoft Exchange 2019 called mail01 and mail02. I recently noticed that browsing to https://mail01.domain.com/ gives a 500 error (well actually it gives Server Error in '/' Application but the logs show the 500 error). However, browsing to https://mail01.domain.com/owa or ecp works fine. Additionally, browsing to https://mail02.domain.com/ also works fine. So it is something in the standard root URL redirection on that one server. The event viewer has the following error: Could not load file or assembly 'Microsoft.Exchange.HttpUtilities, Version=15.0.0.0, Culture=neutral, PublicKeyToken=31bf3856ad364e35' or one of its dependencies. Event code: 3008 Event message: A c

I'm experiencing an intermittent IPv6 connectivity issue on an Oracle Cloud Infrastructure (OCI) instance after migrating my firewall from ufw to nftables. Environment: OS: Ubuntu 24.04 Cloud Provider: Oracle Cloud (OCI) Local Network: Dual-stack (IPv4 and IPv6 available) The Problem: My SSH connection to the server via IPv6 is intermittent, while SSH via IPv4 remains perfectly stable. Upon investigation, I noticed that when the connection fails, the IPv6 address completely disappears from the network interface (checked via ip addr). When the connection is working, the IPv6 address is present. Troubleshooting Steps Taken: I have already allowed all ICMPv6 traffic in the Oracle Cloud Web Console (Security Lists). If I change my local nftables input chain default policy to accept, the IPv6 address reappears in ip addr after a few minutes, and SSH via IPv6 works perfectly again. However, if I change the nftables defa