Why support not responding in 15 hours?

March 27, 2017 1k views
Email DigitalOcean CentOS

Hello,

I've been using digital ocean service for more than 1 year and I am a happy customer of yours. I've also recommended your products to my other friend as well.

I am running a VPS server where more than 100 small websites are hosted.

From yesterday (5:30 GMT) my server is down. By this time I've mailed/respond at least 6 times. But unfortunately I haven't get any response in 15+ hours. I am becoming mad by the phone calls of my clients. :(

What shall I do now? Please help me.

Thanking in advance
Zikrul Ahsan
CEO
[http://www.emythmakers.com/](http://)

4 Answers

Because you're not paying for a SLA. If uptime is important, implement redundancy.

  • Can you kindly give any link for the followings:

    1. What shall I have to do for SLA?
    2. What shall I have to do for redundancy?

    Thanking in advance
    Zikrul Ahsan
    CEO
    [http://www.emythmakers.com/](http://)

      1. It's not offered. 2. Multiple webservers with load balancers.
      • :(. If SLA is not offered yet, what could be other way to get reply for my support ticket?
        About 18 hours has gone. My server is still down. And I am not getting any response yet.

        Please help me or let me know it you have any options?

@emythmakers

Actually, DigitalOcean does provide a 99.99% SLA on Droplets and Block Storage.

https://www.digitalocean.com/help/policy/

The SLA, however, only covers downtime resulting from their end. So if servers fail due to hardware, the network is down, etc -- that's covered.

If the downtime you're experiencing is caused by user-error, being hacked, software error (i.e. issues with Apache, NGINX, MySQL, etc), misconfiguration, resource exhaustion, etc -- that's not covered by the SLA as the service itself is unmanaged and you're responsible for your Droplet and the software running on it including updating, upgrading, security, and optimization. Ultimately, managing a server is your responsibility.

DigitalOcean doesn't have access to your Droplet, thus they can't troubleshoot issues on it beyond asking you to check X, Y, Z and support is limited to what you essentially provide them with. If you need to maintain uptime, redundancy is definitely key.

  • Thanks for your valuable reply dear.

    I think my server is down for one of my sites was attacked by malware. Yes, digital ocean authority mailed my about that issue 2 days ago. But unfortunately I was not in a position to check my email. So I could not fix it in time.

    In your email you also identify where the problem is. Now as the server is down, I cannot access the server and fix the problem. And the problem is unsolved yet.

    Can you kindly give me any suggestion what I can do now?

    • @emythmakers

      If the server is down and you're not able to use SSH to get in, I'd power cycle the Droplet from DigitalOceans control panel to see if that'll bring it back online. To do this, you'll need to login to the control panel, click on the name of the Droplet, then click on Power from the left menu. Power Cycle will be the second option -- click that button.

      At this point, what your main concern should be is gaining SSH access so you can login as that's the only way you're going to be able to figure out what's going on. Without SSH or a way of gaining access to the server, the only resort would be to restore from backups if you have them (i.e. new droplet, new setup, restore data to the new Droplet).

      If you are able to login via SSH after the power cycle, that's when you need to find out what is going on and begin analyzing your server -- check logs, check resource usage, scan for malware, etc.

      If you're not sure how to do that, we can make suggestions, though ultimately, I'd recommend hiring a sysadmin and seeing what they can do to help. This recommendation is based on having owned a few web hosting businesses in the past, knowing the responses you can get from clients, and from working with production data (something you don't want to lose if you can get it back).

      That said, my e-mail is in my profile. I'm available to take a look if you need someone to do so. We can discuss everything over e-mail if you'd like as that'd be more secure when it comes to finer details.

Dear Jonathan,

I've set the power off and on at least 10 times. But still the server is down.

Because of my firewall (cpHulk) I cannot access the server via console or SSH.

I think it is disabled manually from your end. I've already email Digital Ocean email at your email ID.

So, If you kindly enable the server and give me few time to fix it, I think I can do that.

Thanking in advance,

  • @emythmakers

    Just a note, I don't work for DigitalOcean, but I am a sysadmin. I don't have any way to access a Droplet unless it's my own, so I'd need login details to attempt to login to your server (which I'd provide via e-mail, not here in the community).

    That said, looking at the e-mail you sent to me, they've disabled he networking interface on the Droplet, so the only way to access your server at this point would be to use console from within the DigitalOcean control panel. Normal SSH won't work with that interface down.

    You could enable it using console, though that doesn't mean that they won't shut it down again since the issue here is that they've identified phishing attempts coming from your server. That means one of the sites you own, or one of the sites your clients own is a target, or they are in fact phishing.

    This could in fact be due to Malware of some sort, though they only real way to figure this out would be to get them to allow access back and hire someone to fix the issues, or migrate data to a new Droplet (even then, this would still need to be fixed so it doesn't happen again).

    I'll follow up with you via e-mail on this one since access details and such aren't something you need to post here in the community.

Have you checked your DNS provider, because I'm not getting a response from any records on emythmakers.com
Maybe you haven't paid them or their servers are down?

  • @hansen

    Network Interface is disabled in this case, which is why nothing is working :-).

    • @jtittle True dat :) Saw your reply after my post.
      Any of these will result in status: SERVFAIL, so you're not getting far even with a working server.

      dig emythmakers.com A
      dig www.emythmakers.com A
      dig emythmakers.com MX
      

      Unless the NS is being run on the web server too, which is a whole other major issue.

      • @hanesen

        They've disabled any connection other than Console at this point due to the phishing issue.

        Since he's running cpHulk, this is a cPanel server, so the server is handling everything (HTTP/S, DNS, E-Mail, etc).

        With networking essentially disabled, Console is (from their e-mail) the only way they are going to allow access and it's conditional on clean up.

        • I think you are right @jtittle ...
          The DNS server is ok @hansen ..

        • Shall I have to wait for their reply or support in this case or there are other way?

          • @emythmakers

            If you'll check my e-mail, I can get in touch with them, though I need some info to do so. I can't guarantee they'll allow me to access it, but I can try.

          • @emythmakers

            I put in a ticket referencing the ticket ID you provided to see if they would allow me to login, conditional on me disabling access on 80/443 so that any attempts to connect on these ports would be denied until the clean up is done.

            Just replied to your e-mail as well.

            If they will allow temporary access to at least SSH, I can get in and see what sort of issues we're really dealing with. They may need to verify this with you, so bare with me until I hear back from them.

Have another answer? Share your knowledge.