Results 1 to 16 of 16
  1. #1

    Major Problems in New Dedicated Server! Need Help and advice

    Before you read, please understand that this is not a frustration against any company or service but just a request for HELP.

    Ok, Here I am suffering from the migration once more. My Last dedicated server provider hivelocity was excellent in many ways for 3 years. No connection problems, no network issues for 3 years, till it had a a major network attack 2 months back and all my clients had problems of the server falling apart.

    After having server issues like IP attacks, network disconnects, server downtime for more than 2 hrs constantly at times and continued for more than a month, I had to quit on the server. I am sure everyone understands the critical issues when your customers don't get mails and the websites are down.

    I am not a server pro guy. Just a small SEM firm trying to run a small business on a single server with few clients. We don't do spam or any unethical practices, just to let everyone know that we are not spammers, if that's the reason why our servers are getting issues/attacks/downtime.

    Recently I took a server from Proaxxs, the customer service is good, as I did take a small service package too. But satisfaction - I really don't know.

    Here, at WHT I am sure there will be experts who can help me with issues.

    THE RDNS ISSUE
    1) My domain webhostingmonsters on day one was not setup with RDNS and maybe it was my mistake, as I did not ask Proaxxs.net to do a RDNS setup for the domain, which got me blocked out of Yahoo for sending mails from a IP with no RDNS.

    So no mails to yahoo any more. After 3-5 conversations with Yahoo Team on a TS03 ban even after the RDNS propagation, they kept repeating that my domain did not qualify and they wont take off the ban.

    Yes, proaxxs.net helped me do the DKIM authentication for all domains on the server. Still no luck. Any help/advice would be appreciated on getting it fixed.

    NETWORK DISCONNECTS ISSUE
    2) As from Day one, most of clients have been facing a timeout error. Not only customers in India but also in Australia, UK, US too. As many times I have approached the proaxxs team, I have been asked to provide a traceroute from all locations having problem. C'mon how can I ask all my clients to do that?

    Even though some of my clients were helpful and I did send that report to proaxxs team but still the problem remains.

    Even during a cpanel service installation from Configserver.com, they clearly told me that it took them pretty long to install all services due to constant network disconnects on the server. Which was again clearly indicated to proaxxs. But no results till now.

    I had even taken a service for 1 month from websitepulse and created a ping test report every 5 mins from 20 locations in US,UK, Australia, JP and there were more than 30% packet losses from 80% of the locations even in US of which the reports were created and sent to proaxxs.net

    So I am sure, its not just my location. I have had dedicated servers over 4 years now, though not an expert but I can visualize problems like these.

    As an seo sem company we really can't afford the slow speeds of our websites or even network disconnects and now that I see that our online incomes have dropped to 20% since the migration and troubles we have had in the past.

    Al my websites are dead slow, network timeouts are constant. Most of the customers can't send/receive regular emails, yahoo mails is blocked. I think we are in big trouble because I cant figure out where the problem is.

    Do I need an upgrade on something?

    I am still with Proaxxs.net and now searching for the right company to provide me a new server?
    Though my budget is not much but I guess I am really looking now.

    Currently I have a reasonable configuration with a good price
    Core2Quad Q9550 Quad Core Dedicated Server
    Intel(R) Core(TM)2 Quad CPU Q9550 @ 2.83GHz
    8 GB Ram,1000G Hdisk
    With CPanel+fantastico
    + Basic Support
    18 IP's

    As its not easy migrating servers, I would really appreciate a resolution. If no resolution, I would welcome a quote for a good server located in US, east coast preferably with a very stable network.

    Your suggestions will be greatly welcomed
    Thanks in advance.

  2. #2
    Join Date
    Apr 2009
    Location
    Romania
    Posts
    473
    1) The problems with Yahoo are solved a lot faster if proaxxs.net contact them directly.
    But if you do not configure your mailing server right then you will get blocked again.

    2) Check out: http://www.robtex.com/as/as18866.html#graph
    From what I can see proaxxs.net has BGP with 2 internet service providers, but none of them is a Tier 1 ISP.
    PidginHost.com - Managed dedicated servers, cloud servers and software development.

  3. #3
    Thank you ! Well I did tell them about the problem and they said its upto yahoo to resolve this as they have complete right of blocking the shared servers.
    How Do I get them to fix my mailing servers? I mean what's wrong with that?
    If you could kindly clarify.
    Is it the RDNS not set to all the IP's to webhostingmonsters.com?

    Thank you.

  4. #4
    Join Date
    Apr 2009
    Location
    Romania
    Posts
    473
    Some guide lines to follow:
    1. Each account should have a dedicated IP
    2. Automatically send outgoing mail from the account’s IP address instead of the main IP address.
    3. Configure reverse DNS for each IP based on what you have in: /etc/mail_reverse_dns
    4. Remove soft and hard bounce emails based on Yahoo guide lines

    This mailing thing is a bit more complex then you might think, this are the easiest configurations just to ensure that yahoo accepts you.
    PidginHost.com - Managed dedicated servers, cloud servers and software development.

  5. #5
    Join Date
    Mar 2006
    Location
    New Jersey
    Posts
    851
    Naveen, how many websites do you host and what sort of load do you have on the server. It could be that during particular hours some of your websites might be overloading the server to cause timeout.

    As for the emails you must have proper reverse dns, spf record, domain keys and finally request for whitelisting at yahoo, hotmail, aol and other major email service providers.
    24Shells in Business Since 2003 - AS55081
    Dedicated Servers, High Bandwidth Servers
    @24Shells - 24shells.net

  6. #6
    24Shells thanks for asking.
    The apache status during most of the times of network problems is not high at all. The status of CPU and memory used remains below 40%.
    My most of the works are on Joomla, Wordpress and simple HTML's.
    I tested one of the websites on other server by replicating it and transferring the name server for 4 days. Everything worked perfect. Specially the loading time was excellent which I use to have app 2 months back.
    I am checking the logs constantly and cant find any exploits or cpu serge.

    Also if you read my post above, I had taken a server ping and monitoring service from websitepulse.com (amazing service) for pinging the server every 5 mins from 20 locations for 1 month. The packet loss was large.

    If there is any other thing I need to check please do let me know.

    I would appreciate if you could advice if my current RDNS is proper and if not what could I get changed?

    Thank you

  7. #7
    Join Date
    Mar 2006
    Location
    New Jersey
    Posts
    851
    I am not sure what you have currently added but will recommend that you add for eg. server.webhostingmonsters.com on main IP. Than add main IP based SPF on all domains that use shared IP on the server. After 24hours or so of doing this submit a request with Yahoo and explain them that this server is used for shared webhosting.
    24Shells in Business Since 2003 - AS55081
    Dedicated Servers, High Bandwidth Servers
    @24Shells - 24shells.net

  8. #8
    Join Date
    May 2001
    Location
    Around the Block - USA
    Posts
    102
    Naveen,

    We have done everything necessary and possible on our end concerning yahoo. This is a known issue with them and there isn't much more we can do about it, and unfortunately this is now in yahoo's hands.

    On the network disconnects issue at various locations, if you could please update the ticket open in our helpdesk with a "trace route" from a couple of these locations, our admins can take a look. I know you provided a ping report, but we really do need a trace route so we can see where the issue may be. We do not need this from all locations, just one or two that are facing this issue. So, please update the ticket with this info so we can get to the bottom of the issue.

    Thanks!

  9. #9
    Hello Proaxxs,
    I wish I had not to state these problems here. But I have updated the team couple of times with the different trace routes and there has been no improvements.
    I had recently attached a trace route from a customer in Australia about 4 days back. The client still gets the same problem everyday. His last check was yesterday. You can check all my ticket logs.

    But to be honest, what about general website visitors from all locations around the world on various websites? Do I ask them for a traceroute too?

    It's more scary to imagine everyday, which part of the world can't open our websites. Once in doubt, always a doubt if not resolved in first few days/weeks.

    I have no doubt in your services, they are excellent, but sorry to say the network is pathetic.

    Thanks for asking again, because that's the only thing happening since day one on this issue. You have been asking and I have been replying constantly.

  10. #10
    Join Date
    May 2001
    Location
    Around the Block - USA
    Posts
    102
    I'm looking at your ticket Naveen and you provided one tracert, but provided several ping reports. You still have not updated the active ticket for 2 days. Please update your ticket for proper support. A forum is not the proper way to provide support.

    To say our network is pathetic is really unfair. If we do not see the problem how can we help? A trace should give us an indication of where the issue is. Again I urge you to update the ticket otherwise there isn't much we can do.

  11. #11
    Sure. Now I shall update with more Trace routes, before we come to next conclusion.
    Though there have been 3 trace routes provided in last 1.5 months.
    The last trace route was mentioned here. Even that location had problems 2 days back.
    I shall update all trace routes again as in any case my first choice is not moving out at some problem that can not be solved.

    In the coming week now I shall update all locations with problems to give trace routes if possible.

  12. #12
    Quote Originally Posted by NaveenKapur View Post
    The apache status during most of the times of network problems is not high at all. The status of CPU and memory used remains below 40%.
    the apachestatus cpu and memory use figures are always inaccurate the point of uselessness. From ssh shell / putty, try "top -n 1" and "free -m" and paste those here, as those give a much more accurate idea of the load on the system. You could easily be using all your ram and swap and apache status will still happily report 20-40% memory use.
    IOFLOOD.com -- We Love Servers
    Phoenix, AZ Dedicated Servers in under an hour
    ★ Ryzen 9: 7950x3D ★ Dual E5-2680v4 Xeon ★
    Contact Us: sales@ioflood.com

  13. #13
    Hi NaveenKapur,

    Is your issue resolved?

    May I know the IP of your server? Or do you know which data center is your server located?

    I am considering getting a new server from them. If possible I'll avoid the data center.

    Thanks.

  14. #14
    Join Date
    May 2001
    Location
    Around the Block - USA
    Posts
    102
    It has been confirmed this is NOT a data center issue. I understand it is always easier to blame the data center for connection issues, but this is not always the case.

    Why this happens from different locations is most likely a dns issue and not an issue on our end.

    On the trace ran to the server there is already a lost packet and high latency right when it leaves the connection.

    On the way back to workstation the connection is fine until it reaches the last hop and then the latency shoots way up.

    This issue is far out of our network influence and there is nothing we can do to repair it. In fact it seems like it may be the provider since they are the first and last hops of both routes. Usually when providers have these sorts of latency issues they repair them within 24 to 48 hours.

    In fact, just to be on the safe side we asked cPanel techs to take a look. They confirmed the same.

    on host-tracker though when you pinged the domain name, some servers did not resolve the domain name correctly it seems and got errors. This can be due to the cache in certain name servers. Some might not have updated your domain name correctly. Or perhaps there are issues with some DNS settings that can be blocking some IPs and thus not resolving the name.

    Naveen, your ticket was updated in regards to our findings.


    I would just like to say that if it's something on our end, we will accept the blame, but trying to make us look bad for something out of our control is really unjustified. But I suppose that's the nature of the beast.

  15. #15
    Thank you for the report. I see the sites are performing better than before. And the good part is I have not had 1 complaint from a single customer in last 4 days compared to what I was getting before. (Every 2nd day)

    But please do correct me if i am wrong.
    I see all this happening after I asked your support team to check on IP optimization and some core reports they gave me.

    Do you think its because of the server IP optimization?

    I mean you can't all of a sudden expect not receiving anymore disconnects or mail blocks from all who complained for more than a month.
    I personally have been calling my clients, chatting and emailing them to know about the current network status since 2 days. And everybody has clearly told me that the service is NOW better.

    If it was a DNS issue, it just did not get solved in last few days after my posting a problem in WHT, did it?

    So, just wanted to know, what did you do?

    But thanks for what you did. You have been supportive. I wish that this problem was not brought here to get things sorted.

  16. #16
    After many email interactions, I believe it was really a DNS problem and not a proaxxs network problem.

    Only if I had been given a detailed analysis by Proaxxs before-hand, of the extra man hours and investigations on my complaints spent by them on resolving this issue, I would have probably never complained.

    But I believe its part of their good service on not letting the customers know how much extra efforts they put in resolving such issues for each customer.

    I am not saying they are perfect, but then no-one is.

    After looking at the reports, I think I ll stay for now with Proaxxs and give it another shot.

    Thank you Proaxxs

Similar Threads

  1. having major nameserver/server/domain problems...need big help
    By mikey1090 in forum Programming Discussion
    Replies: 3
    Last Post: 02-15-2007, 06:30 AM
  2. Major server performance problems after "upgrade"
    By slack in forum Hosting Security and Technology
    Replies: 11
    Last Post: 03-10-2006, 12:52 AM
  3. Major, MAJOR Problem With Server - Need Advice!
    By CISupport in forum Hosting Security and Technology
    Replies: 23
    Last Post: 11-21-2004, 09:51 PM
  4. Having major server problems
    By DigiCrime in forum Hosting Security and Technology
    Replies: 6
    Last Post: 05-14-2003, 03:47 AM
  5. Major Server Problems
    By tymonhall in forum Dedicated Server
    Replies: 3
    Last Post: 10-26-2001, 12:18 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •