Page 2 of 7 FirstFirst 1234 ... LastLast
Results 11 to 20 of 64
  1. #11
    Master Untangler neiby's Avatar
    Join Date
    Jun 2009
    Location
    Denver, CO
    Posts
    603

    Default

    Arg! It's doing it again!! I really need to figure this out fast or I'm going to have to undo everything before I leave. We can't have our websites unavailable, especially every 10 minutes.

    i've noticed that every time it has locked up, I've been in the Protocol Control settings page. It seems to lockup when I switch pages. Protocol Control is not turned on yet. I'm just going through and configuring it in preparation to turn it on.
    Disclaimer: I may or may not have had enough coffee when I'm posting. Interpret my responses thusly.

  2. #12
    Master Untangler neiby's Avatar
    Join Date
    Jun 2009
    Location
    Denver, CO
    Posts
    603

    Default

    Come to think of it, I think every time it has locked up like this I was in the Protocol Control settings page. That can't be a coincidence, can it? The rack unit isn't even turned on, so I don't know why that would cause a problem.

    I'm going to stay out of it and see if it locks up again. It seems to be locking up for about 10 minutes every 10 minutes.
    Disclaimer: I may or may not have had enough coffee when I'm posting. Interpret my responses thusly.

  3. #13
    Master Untangler neiby's Avatar
    Join Date
    Jun 2009
    Location
    Denver, CO
    Posts
    603

    Default

    It's doing it again and I wasn't on the Protocol Control settings page. Grr.... I'm going to have to take UT back offline. Darn it. I thought we had it working.
    Disclaimer: I may or may not have had enough coffee when I'm posting. Interpret my responses thusly.

  4. #14
    Master Untangler neiby's Avatar
    Join Date
    Jun 2009
    Location
    Denver, CO
    Posts
    603

    Default

    Okay, time for some troubleshooting. When in production, ever ten minutes or so, the web UI becomes unresponsive. During that same period, outside users can not get to our website and employees cannot get to their Outlook Web Access pages, which also messes up email for users with iPhones.

    Regular outbound Internet traffic doesn't seem to be affected at all during this period. I checked our firewall to see the number of sessions in use and it's not that high. It's probably higher right now than it was last night when the problems were occurring. It shows 1012 current connections with 2827 the maximum. That includes TCP and UDP connections.

    Any idea where to start troubleshooting?

    Many thanks!
    Disclaimer: I may or may not have had enough coffee when I'm posting. Interpret my responses thusly.

  5. #15
    Master Untangler neiby's Avatar
    Join Date
    Jun 2009
    Location
    Denver, CO
    Posts
    603

    Default

    Oh, I forgot to mention something that I think may be important. We have a DNS server on our DMZ that is authoritative for our domain. That server only responds to queries for sites in our domain. It will return publicly routable addresses for those sites. We have a separate internal DNS server for our organization. If you resolve the names of our websites from a PC on the inside network, you will get an internal private IP address, the real address of the server.

    Since (other than the UT web UI) the only things that seem to be affected during this outage are our sites visible from the outside world, I suppose this could point to a DNS problem.

    When the problem began, I had UT pointing to our internal DNS servers. Dmorris suggested this might be a DNS problem, so I pointed UT to OpenDNS instead. The problem is still occurring. Even if this isn't a DNS problem, what would be better, to point UT to external or internal DNS? Or does it matter as long as it responds?

    Thanks!
    Disclaimer: I may or may not have had enough coffee when I'm posting. Interpret my responses thusly.

  6. #16
    Untangle Ninja sky-knight's Avatar
    Join Date
    Apr 2008
    Location
    Phoenix, AZ
    Posts
    26,482

    Default

    That is a great big IT DEPENDS!

    I try to keep my UT configured the same as the clients on the segment so they both resolve things the same way.

    But, my largest install is configured with OpenDNS, and the AD servers are configured to forward to UT. So the chain flows outward...

    However, that does mean the UT itself can't see the split internal namespace entirely, and because of that cannot send e-mail directly via the exchange server there.
    Rob Sandling, BS:SWE, MCP
    NexgenAppliances.com
    Phone: 866-794-8879 x201
    Email: support@nexgenappliances.com

  7. #17
    Master Untangler neiby's Avatar
    Join Date
    Jun 2009
    Location
    Denver, CO
    Posts
    603

    Default

    Quick update. We discovered that outside users ARE able to get to the website on the DMZ. That was really hurting my head trying to figure out why that would stop working!

    So, the problem now is just that users intermittently can't get to OWA. We're testing it now and it was working for a few minutes and now it's being blocked. As before, the UT web UI is locked up, too.

    We've been monitoring the firewall. There really aren't that many connections. There are 1181 current connections. That's nothing. And on UT, we have 1.7 GB of RAM free and the CPU usage is very low.

    I'm not sure where to look. I wonder if a tcpdump on UT would help. I'll have to refresh my memory. I haven't used tcpdump in a couple of years. If I can limit it to recording traffic to our OWA server, that would be awesome.
    Disclaimer: I may or may not have had enough coffee when I'm posting. Interpret my responses thusly.

  8. #18
    Master Untangler neiby's Avatar
    Join Date
    Jun 2009
    Location
    Denver, CO
    Posts
    603

    Default

    Okay, I just found a really weird clue. I was looking in the eSoft web filter logs. Nothing related to our internal site was in the blocked logs, so I looked in the passed logs. I see entries from the outside world coming into our OWA site. The log says it was allowed to pass, but then over in the "reason" column it says "in Categories block list".

    Weird! I'm including a screen shot. It's almost as if the web filter thinks the traffic is being passed, but at the same time thinks that our internal site is in the categories block list. I'm including a screenshot.

    I've disabled the web filter to see if the problem goes away. If it does, that narrows it down a bit.
    Disclaimer: I may or may not have had enough coffee when I'm posting. Interpret my responses thusly.

  9. #19
    Master Untangler neiby's Avatar
    Join Date
    Jun 2009
    Location
    Denver, CO
    Posts
    603

    Default

    We verified that this does not happen if the eSoft web filter is disabled. It appears to be intermittently blocking access to our internal OWA server from the outside. That's weird enough, but then it logs it on the "passed" logs screen. And why would the web UI lock up? So odd...

    I've turned eSoft back on for more testing.
    Disclaimer: I may or may not have had enough coffee when I'm posting. Interpret my responses thusly.

  10. #20
    Master Untangler neiby's Avatar
    Join Date
    Jun 2009
    Location
    Denver, CO
    Posts
    603

    Default

    Now we have a situation where access to OWA is okay, but the web UI is still locking up. Maybe those two events are unrelated. Something keeps locking up the UI, though. Perhaps it's unrelated to what is blocking access to OWA.
    Last edited by neiby; 07-15-2009 at 09:01 AM.
    Disclaimer: I may or may not have had enough coffee when I'm posting. Interpret my responses thusly.

Page 2 of 7 FirstFirst 1234 ... LastLast

Posting Permissions

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

SEO by vBSEO 3.6.0 PL2