Incidents don't announce themselves and they don't keep office hours. A server stops responding at 3 a.m., a certificate renewal fails on a Sunday, an attack saturates a link over a weekend. That is when the quality of a support team is actually measured, far more than in any sales brochure.
At Infrawire, technical support is reachable 24 hours a day, 7 days a week, and is aimed at technical users. This article explains what that implies, what you can reasonably expect from a hosting provider's support, and how to word a request so it gets handled without needless back-and-forth.
Why round-the-clock availability matters
Hosted workloads don't follow a calendar. A trading VPS is busy during the Asian and US market sessions, an online shop sells in the evening, an API serves customers across several time zones. If support only answers from 9 to 6, a problem that starts on Friday evening waits until Monday.
What 24/7 mainly reduces is uncertainty. Knowing that a technical contact exists at any hour lets you decide more calmly: wait, try a fix yourself, or escalate. It doesn't remove the need to monitor your own services, but it keeps you from facing an infrastructure problem alone.
What "expert" should mean in practice
First-line support that points you to a FAQ is useless when the problem is an inconsistent routing table or a failing disk. Expert support starts with someone who reads your whole message, understands which layer is involved (network, hypervisor, system, application) and answers with a diagnosis rather than a stock phrase.
A few signs that rarely lie:
- answers cite verified elements (interface, route, log, machine state) instead of generalities;
- the support team can tell an infrastructure problem from a client-side configuration problem, and explains the difference;
- when a cause is identified, you get the reason, not just "it's fixed".
What support covers and what stays on your side
The exact scope depends on the plan you choose, and it is better to read it before you need it. As a general rule, an unmanaged VPS or dedicated server splits responsibilities like this.
| Topic | Provider side | Customer side |
|---|---|---|
| Hardware, hypervisor, power | Yes | No |
| Network and connectivity of the machine | Yes | Checking local rules |
| Console access, reboot, reinstallation | Yes, depending on the plan | Request and prior backup |
| Firewall and service configuration | Advice | Yes |
| Application code, database | No | Yes |
This boundary is normal. Support can't debug your business application, but it can help you show that the problem doesn't come from the machine. For volumetric attacks, a dedicated offering such as the Anti-DDoS VPS addresses a different need than a standard VPS.
Preparing your request to save time
The difference between a ticket resolved in one exchange and one that takes five almost always comes down to the information provided up front. Before writing, gather a few things.
Describe the symptom and the time
State what stopped working, since when (with the time zone) and what changed just before: an update, a firewall change, a new deployment. "The site is slow" is unusable. "Requests to port 443 have been timing out since 22:10 UTC, with no configuration change" is not.
Attach diagnostics
For a network issue, a trace in both directions is the most helpful thing you can provide. You can also compare with the looking glass to check reachability from the outside.
Bashmtr -rwzc 100 votre-serveur.exemple.com
For a system issue on Linux, these commands give you the essentials:
Bashjournalctl -p err -b --no-pager | tail -n 50 df -h free -m ss -tlnp
For a Windows VPS, say whether the problem affects RDP or the service itself, and include what Event Viewer shows.
Say what you have already tried
A reboot, a disabled firewall rule, a test from another network: this information saves repeating the same checks and points straight to the right layer.
How to evaluate a hosting provider's support
Before choosing, ask two or three concrete questions rather than relying on a bare "24/7" mention. Does support answer a precise technical question before purchase? Is the scope clearly written down? Is there documentation you can use on your own? The Infrawire tutorials exist precisely so you can solve common cases yourself, with support taking over for the rest.
Also test with a real, low-risk case: a question about network configuration or reinstalling a system. How you are answered before an incident says a lot about what will happen during one.
If you are comparing plans, the VPS catalogue specifies what each plan includes, and it is the reference page for the applicable conditions.
FAQ
No. Support works on the infrastructure and advises you on configuration, but it doesn't watch your applications for you. Set up alerts on your services (HTTP, database, disk space) so you are warned before you need to open a ticket.
It depends on the plan. On an unmanaged server, administering the system and applications is your responsibility, while support focuses on the infrastructure and on helping with diagnosis. Check the scope of the plan before ordering.
Back up your data off the machine, because a reinstallation wipes the disk. In your request, state the operating system and version you want, along with your SSH keys or any access requirements.
Yes, a clear one. A dated symptom, command output and a list of what has already been tested let support go straight to diagnosis without asking preliminary questions.