Results 1 to 12 of 12
  1. #1

    Pegasus Web Tech / Nac uptime

    Hi,

    I've got a server inside NAC Parsippany with Pwebtech on their DCJN network. I decided I would try and work out my uptime for the year so far. The most reliable monitoring I have on it is openacs uptime http://uptime.openacs.org/ which requests a page from your http server every approx 15 minutes. So there is probably more reliable monitoring that others are doing. The server has been rebooted once (or twice?) and apache restarted a few times - but nothing that would cause the server to appear "down".

    Anyway, I work out my uptime using openacs uptime to be:

    99.995%

    Anyone want to compare their NAC results? I think DCJN was having some router/routing trouble earlier in the year.

  2. #2
    Join Date
    Feb 2002
    Location
    Australia
    Posts
    24,009

    Re: Pegasus Web Tech / Nac uptime

    Originally posted by freakysid
    Anyway, I work out my uptime using openacs uptime to be:

    99.995%
    You should be able to do better than that!!
    AussieHost.com Aussie Bob, host since 2001
    Host Multiple Domains on Fast Australian Servers!!

  3. #3
    Join Date
    Nov 2001
    Location
    Canada
    Posts
    1,963
    well i dont have any monitoring like that setup to log percentage

    but i can say that my servers at nac (207.99.115.* ) have been down twice last month

    one lasted 6+ hrs, and the other lasted i believe 30 mins tops

    aside from those two episodes, no down time what so ever

  4. #4
    Originally posted by clocker1996
    well i dont have any monitoring like that setup to log percentage

    but i can say that my servers at nac (207.99.115.* ) have been down twice last month

    one lasted 6+ hrs
    Ouch! Sh1t happens I guess.

    I worked out the percentage myself with a bit of exel voodoo. Here are the raw stats for this year (remember the httpd server is polled every 15 mins approx so the downtime is not accurate to the minute, and some very minor outages may not be reported at all).

    2002-05-11 21:17:31 : back_up
    2002-05-11 20:49:04 : down
    2002-04-25 16:53:19 : back_up
    2002-04-25 16:24:51 : down
    2002-04-23 20:19:05 : back_up
    2002-04-23 19:53:10 : down
    2002-04-23 04:30:54 : back_up
    2002-04-23 04:05:30 : down
    2002-04-23 02:24:04 : back_up
    2002-04-23 01:53:49 : down
    2002-04-09 05:32:14 : back_up
    2002-04-09 05:10:05 : down
    2002-04-05 03:33:54 : back_up
    2002-04-05 03:06:41 : down
    2002-04-04 00:13:45 : back_up
    2002-04-03 23:45:34 : down
    2002-04-03 23:20:38 : back_up
    2002-04-03 19:53:04 : down
    2002-03-26 20:31:07 : back_up
    2002-03-26 19:40:26 : down
    2002-03-26 08:33:23 : back_up
    2002-03-26 07:44:58 : down
    2002-03-21 01:35:09 : back_up
    2002-03-21 00:57:11 : down
    2002-03-14 14:28:58 : back_up
    2002-03-14 13:38:33 : down
    2002-03-05 10:12:24 : back_up
    2002-03-05 09:46:45 : down
    2002-03-02 13:56:45 : back_up
    2002-03-02 13:30:22 : down
    2002-02-25 09:47:17 : back_up
    2002-02-25 08:57:08 : down
    2002-02-24 06:04:37 : back_up
    2002-02-24 05:34:15 : down
    2002-02-24 00:50:24 : back_up
    2002-02-24 00:24:12 : down
    2002-02-23 21:03:52 : back_up
    2002-02-23 20:33:18 : down
    2002-02-21 17:20:14 : back_up
    2002-02-21 16:51:53 : down
    2002-02-17 17:02:35 : back_up
    2002-02-17 16:23:12 : down
    2002-02-16 21:28:56 : back_up
    2002-02-16 20:31:47 : down
    2002-02-15 15:31:11 : back_up
    2002-02-15 14:49:30 : down
    2002-01-27 23:46:05 : back_up
    2002-01-27 15:11:52 : down
    2002-01-27 12:43:20 : back_up
    2002-01-27 12:15:39 : down
    2002-01-23 00:36:36 : back_up
    2002-01-22 12:00:32 : down
    2002-01-20 15:48:27 : back_up
    2002-01-19 09:58:41 : down
    2002-01-17 22:07:55 : back_up
    2002-01-17 21:31:00 : down
    2002-01-15 03:11:47 : back_up
    2002-01-15 00:37:58 : down
    2002-01-14 17:27:02 : back_up
    2002-01-14 12:47:22 : down
    2002-01-08 14:33:55 : back_up
    2002-01-08 14:05:04 : down

    I'm on the 64.247.* block

  5. #5
    Just to be clear your sever is based out of Newark NJ, not parsippany. And the reason you were down was not becuase of NAC but becuase of a failed cross connect that equinix installed. Also clocker is not a pwebtech client.

    Freakysids uptime does represent a good estimate of pwebtech/nac in parsippany uptime.

    Originally posted by clocker1996
    well i dont have any monitoring like that setup to log percentage

    but i can say that my servers at nac (207.99.115.* ) have been down twice last month

    one lasted 6+ hrs, and the other lasted i believe 30 mins tops

    aside from those two episodes, no down time what so ever
    Jay

  6. #6
    Originally posted by freakysid


    Ouch! Sh1t happens I guess.

    I worked out the percentage myself with a bit of exel voodoo. Here are the raw stats for this year (remember the httpd server is polled every 15 mins approx so the downtime is not accurate to the minute, and some very minor outages may not be reported at all).

    2002-05-11 21:17:31 : back_up
    2002-05-11 20:49:04 : down

    ...

    Unless I misunderstood, the problem with those type of stats, are that any web server issues will throw off the results. Syn attacks on the web server port, a slow response, overloaded web server, or anything. Of course it's not worthless, but there's better ways to monitor a server than just web server/httpd polling and those would likely be more accurate.
    Robert McGregor
    URL: http://www.2host.com
    Email: robertm@(nospam)2host.com

  7. #7
    Join Date
    Oct 2001
    Location
    God's Country :)
    Posts
    760
    We have 2 Servers currently with Pegasus / DJCN and have had no downtime except a few flickers which wasn't down just a normal network thing. Clients are very happy with the speed they're getting from NAC. Pegasus also just now has an AT&T line which was installed today. From Kentucky I'm steady getting a 60 - 72 ping time which is excellent from my distance away from NJ. Jason has also been very helpful when we needed something. I also always liked Ed me and him have had quite some conversations on the phone. Our monitors are showing 104.2339% Uptime.

  8. #8
    I know about down time! I had two servers on BurstNet!

    I have not had any down time on Pegasus at all!

    I moved our first server over about 80 days ago. It ran with out a reboot for 69 days. I guess they rebooted to get the 2nd server ready to move off BurstNet last Friday!

    In the time frame we have been on Pegasus their network is rock solid!

    After coming from "DDoSBurstNet" Pegasus is uptime heaven.

    Regards,
    Dale Crow
    GH1.com

    Be looking for the new noval "The Tail of Two BurstNet Servers" coming to a WebHostingTalk Forum near you! I am waiting on BurstNet to write the last chapter before posting it. The last chapter will be billing and miscellaneous over Bandwidth charges!
    Dale Crow
    GH1 Web Hosting, Inc.
    http://GH1.com
    The attention span of a computer is only as long as its power cord.

  9. #9
    Originally posted by 2host.com


    Unless I misunderstood, the problem with those type of stats, are that any web server issues will throw off the results. Syn attacks on the web server port, a slow response, overloaded web server, or anything. Of course it's not worthless, but there's better ways to monitor a server than just web server/httpd polling and those would likely be more accurate.
    That's a fair call. Those stats will reflect not just nac / pwebtech's performance but my performance too as the sysadmin. However, I also think from a shared hosting customer's perspective, the effective uptime of the httpd is important because essentially that's what they will be looking at (and that is what they will be using to monitor their virtual server) - that ppl can reach their web site.

    Also, those stats will be polluted by any network trouble at openacs.org end of the network too (as it is only one location that is polling the httpd).

    I thought it would be fun to try and calculate an uptime from the data I do have - and because I know there were some troubles with DCJN earlier in the year, I am pleasantly suprised with the result. Note, no downtime over the past three months. While we always have a subjective rought idea of these things, it is nice to step back and try and see how that fits with the objective data on hand.

  10. #10
    One thing you can do to maybe balance that out, is have some local accesses to the web server I.e., in the 127.x.x.x range and see if it's up or responsive to a reasonable degree. That will rule out network issues for the stats and you can better determine where the actual problem lies. I definitely agree that wherever the problem lies (be it solely or in whole/combined) is irrelevant if the service(s) are/is down. Perhaps that would help, but there's likely many tools/scripts out there for that task.
    Robert McGregor
    URL: http://www.2host.com
    Email: robertm@(nospam)2host.com

  11. #11
    Join Date
    Feb 2002
    Location
    Australia
    Posts
    24,009
    Originally posted by TimPD
    Our monitors are showing 104.2339% Uptime.
    AussieHost.com Aussie Bob, host since 2001
    Host Multiple Domains on Fast Australian Servers!!

  12. #12
    Join Date
    Apr 2002
    Location
    Auckland - New Zealand
    Posts
    1,572
    104.2339% Uptime.
    Well the saying "There just are not enough hours in the day" rings true with that statement - TimPD 100% is the maximum achievable - I think that your uptime monitor may need a tweek,


Posting Permissions

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