Results 1 to 11 of 11
  1. #1

    Trust in bandwidth usage graphs

    Hi all

    I have a question about bandwidth usage graphs which I think many UK based co-location providers for example, provide their customers so they can monitor their usage.

    If a server host provides a highly inconsistent bandwith usage graph, which happens to show ones bandwidth dropping massively in coindidence with a hardware changed they announced, is there anything that can be done about this? For example if the host in question upgraded their DNS server.

    What would be your recommended action/ advice. The scenario you would have is being charged a much higher monthly bill than necessary for bandwidth which will go un-used. The previously inaccurate graphs influenced you to going to the higher bandwidth monthly plan in the 1st place.

    If they provide bandwidth usage graphs, and it was found that bandwidth measurements were previously distorted- should the hosting provider have a duty at least to adjust ones plan to the one more appropriate - without any hassle or strings attached. Also in the extreme - if it is found to be their fault for previously inaccurate graphs/ grossly mis-represented bandwidth, then should they actually refund money?

    Help and advice appreciated

    Best wishes
    Tryf

  2. #2
    Join Date
    Jul 2003
    Posts
    277
    So because you had less traffic 'the theoretical upgrade to the DNS Server', you bought MORE bandwidth? And if it's colo, I'd run mrtg or something of the sort on my own box, to show how much is Actually going through the box.
    Perl
    Java
    SQL

  3. #3
    Join Date
    May 2003
    Location
    Port Arthur, Texas
    Posts
    1,650
    Is MRTG/SNMP [or whatever graphing program you are using] running on the switch port or on your machine?
    Become an ISP; a great value-added service for any web host.
    http://www.vispprovider.com

  4. #4
    Join Date
    Jul 2003
    Location
    London UK
    Posts
    327
    Graphing isn't a great way of determining usage for billing purposes IMO. More useful for showing you trends, ie. peaks and usage patterns.

    For this reason we chose to provide monthly/live bytes in/out counters as well as graphing.
    Paul Civati
    Rack Sense Ltd UK Managed Services Provider
    Views expressed are my own and not those of the company.

  5. #5
    Join Date
    Jul 2002
    Location
    London, United Kingdom
    Posts
    4,212
    If the host has been providing *proven* inaccurate monitoring of your bandwidth, then yes, you should have some form of comeback

    Whether the host wants to retain you as a customer will determine how well they act, although I would always advocate running your own graphing and byte-counts for verification/validation anyway.
    Rob Golding Astutium Ltd - UK based ICANN Accredited Domain Registrar - proud to accept BitCoins
    Buying Web Hosts and Domain Registrars Today @ hostacquisitions.co.uk
    UK Web Hosting | UK VPS | UK Dedicated Servers | ADSL/FTTC | Backup/DR | Cloud
    UK Colocation | Reseller Accounts | IPv6 Transit | Secondary MX | DNS | WHMCS Modules

  6. #6
    Join Date
    Apr 2003
    Location
    california
    Posts
    152
    the graphs can be deceiving at times.

    go for another method as well as MRTG, etc.

    I always like to be safe rather than sorry and get more bandwidth though

  7. #7
    Join Date
    Apr 2003
    Location
    Vancouver, Canada
    Posts
    760
    Install MRTG locally to monitor your bw usage. Some providers sometimes don't have your monitoring set to your port, etc.
    Paradox Web Hosting - codyw@paradox-group.com
    - Shared and Dedicated Hosting Solutions: www.paradoxwebhosting.com
    - Auto Sales Manager 2004 now available!: www.autosalesmanager.com

  8. #8
    I agree with installing MRTG on your own server as it is the only sure way (if configured correctly of course) to ensure the bw usage is correct.
    Host Byte Marketplace
    Directory of Web Hosting Companies, Services and Resources
    http://www.hostbyte.com
    Add Your Web Hosting Product Today!

  9. #9
    Join Date
    Jul 2002
    Location
    London, United Kingdom
    Posts
    4,212
    at least add some form of traffic count over the interface - mrtg is *not* the only method to measure b/w but has it's uses ..
    Rob Golding Astutium Ltd - UK based ICANN Accredited Domain Registrar - proud to accept BitCoins
    Buying Web Hosts and Domain Registrars Today @ hostacquisitions.co.uk
    UK Web Hosting | UK VPS | UK Dedicated Servers | ADSL/FTTC | Backup/DR | Cloud
    UK Colocation | Reseller Accounts | IPv6 Transit | Secondary MX | DNS | WHMCS Modules

  10. #10
    True!

    Even better if you hook it up into your own router..but I doubt its required in this case.
    Host Byte Marketplace
    Directory of Web Hosting Companies, Services and Resources
    http://www.hostbyte.com
    Add Your Web Hosting Product Today!

  11. #11

    Bandwidth graphs

    Hi all

    I hope this helps anyone who managed to create excess bandwidth the way I did in an extremely dramatic fashion. I have a popular Email chess server called Chessworld.net which sends out Emails every second. When a player moves it sends out an Email to the opponent to indicate they have moved.

    The hosting provider upgraded a DNS server at one of their datacenters and I noticed that coincidentally this impacted my bandwidth graph - instead of peaking at 200/300 bps it was less than 100bps peaks for over a week. I.e. instead of approaching 75 Gig a month, it was now less than 30 gig a month(!).

    I began investigating the DNS set up on my server - a caching only DNS - and it turns out that we should have set up forwarders, and also TURNED OFF RECURSION. Why on earth is recursion the default?! I am not a DNS guru, but if your webserver sends out a tonne of emails every day, it is worth experimenting with your DNS set up.

    We did full DNS logging but without the last two fields, and there were tonnes of excess DNS related stuff being logged. By taking off recursion, when sending Emails, it started doing just cached DNS lookups from our server. It was now not using recursion with loads of different navigations to try and send an Email to someone who probably put a dodgy Email address in anyway.

    I think that their DNS upgrade probably dramatically lessened the DNS related traffic - and gave me the clue to start investigating any potential improvements in our own DNS set up. I believe the setup is now much better with forwarders put in, and recursion switched off. Additionally the retry count on Emails in the SMPT set up set to maximum 2 retries - a time out of one hour on non-delivery. Emails and their related traffic were the cause of my bandwidth woes.

    Thank you for all that posted in this thread.

    Best wishes
    Tryfon

  12. Newsletters

    Subscribe Now & Get The WHT Quick Start Guide!

Related Posts from theWHIR.com

Posting Permissions

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