Page 2 of 3 FirstFirst 123 LastLast
Results 11 to 20 of 27
  1. #11
    Untangler
    Join Date
    Jun 2008
    Posts
    55

    Default

    Quote Originally Posted by sky-knight View Post
    Building a new VM might not solve the problem, but it should rule out any potential corruption of the install itself.

    But, don't discount that drive IO just yet. Is the mail server Untangle is protecting on the same set of disks? Because regardless of configuration a busy mail server, with an anti-spam engine that lives and dies by eavesdropping such as Untangle, on the same disks?!? That's how you create race conditions... I'm not saying that's what's happening, I'm just saying there be dragons here. IO load in these configurations is highly contingent on mail volume, and you may have blobs of SMTP traffic landing at once and it's becoming a DOS. These things aren't as clear when you run a virtual Untangle, if it was bare metal you'd see it far more easily.
    Well, I'm able to see very granular metrics for the Untangle VM (e.g. Avg. IOPS 42, Peak IOPS 169), and the overall metrics for the flash array are almost always sub millisecond and less than 2,000 IOPS (25,000 IOPS Capacity).

    I'm going to try and build a new Untangle box today and tomorrow and see what happens.

    @dmorris, with regards to Untangle support I've kind of run into a dead end. Collen has stopped responding to my calls and ticket updates.

  2. #12
    Untangle Junkie dmorris's Avatar
    Join Date
    Nov 2006
    Location
    San Carlos, CA
    Posts
    17,747

    Default

    What is your ticket #?
    Attention: Support and help on the Untangle Forums is provided by volunteers and community members like yourself.
    If you need Untangle support please call or email support@untangle.com

  3. #13
    Untangler
    Join Date
    Jun 2008
    Posts
    55

    Default

    #118244 Thanks!

  4. #14
    Untangle Junkie dmorris's Avatar
    Join Date
    Nov 2006
    Location
    San Carlos, CA
    Posts
    17,747

    Default

    Are you sure? The ticket details a wildly different scenario than what you described above.
    In fact, you clearly state that "bypass SMTP traffic on Untangle the problem goes away completely"

    If your mail server is sending a 4.4.1 timeout you must increase the timeout on your mail server, or you must at least figure out why it is sending it. Untangle is not going to send an exchange error message. Untangle does not run exchange.
    The fact that you assert "It must be an issue with Untangle because the issue disappears when I remove Untangle" is not true.
    Clearly the timeout requirements are much higher with Untangle in-line because the scans must take place.

    Support can help you and give you guidance, but ultimately you are free to do as you choose. If you choose to disregard their advice and do not want to troubleshoot because you believe the issue is elsewhere, there is not much they can do.

    I would also check your basic settings like DNS to make sure scans don't take super long because DNS is not working or something obvious like that. However, we see this issue fairly regularly and it happens anytime the timeouts on the mail server are low.

    (And no, reinstalling Untangle won't change anything about the above unless you modified the timeouts within untangle or other default params that resetting might help)
    Last edited by dmorris; 04-03-2018 at 01:02 PM.
    Attention: Support and help on the Untangle Forums is provided by volunteers and community members like yourself.
    If you need Untangle support please call or email support@untangle.com

  5. #15
    Untangle Ninja sky-knight's Avatar
    Join Date
    Apr 2008
    Location
    Phoenix, AZ
    Posts
    23,774

    Default

    Quote Originally Posted by soccerextreme View Post
    Well, I'm able to see very granular metrics for the Untangle VM (e.g. Avg. IOPS 42, Peak IOPS 169), and the overall metrics for the flash array are almost always sub millisecond and less than 2,000 IOPS (25,000 IOPS Capacity).
    I agree, that doesn't seem stressed at all, which either means your VM is corrupt or the hypervisor is really screwing up disk scheduling for some reason. That is assuming some not so obvious configuration fault is to blame.

    I'd give UT support some more time, Collin is a great guy but he's far from all knowing! And now you've gained the attention of the all mighty DMorris... Which as much as I know it pains him for me to say this, but if he make progress on this no one can!
    Rob Sandling, BS:SWE, MCP
    NexgenAppliances.com
    Phone: 866-794-8879 x201
    Email: support@nexgenappliances.com

  6. #16
    Untangler
    Join Date
    Jun 2008
    Posts
    55

    Default

    Quote Originally Posted by dmorris View Post
    Are you sure? The ticket details a wildly different scenario than what you described above.

    If your mail server is sending a 4.4.1 timeout you must increase the timeout on your mail server, or you must at least figure out why it is sending it. Untangle is not going to send an exchange error message. Untangle does not run exchange.
    The fact that you assert "It must be an issue with Untangle because the issue disappears when I remove Untangle" is not true.
    Clearly the timeout requirements are much higher with Untangle in-line because the scans must take place.

    Support can help you and give you guidance, but ultimately you are free to do as you choose. If you choose to disregard their advice and do not want to troubleshoot because you believe the issue is elsewhere, there is not much they can do.

    (And no, reinstalling Untangle won't change anything about the above unless you modified the timeouts within untangle or other default params that resetting might help)
    Yes, that is the correct ticket. The 4.4.1 timeout message is generated by the sender's server not the recipient server.
    Last edited by soccerextreme; 04-03-2018 at 12:58 PM.

  7. #17
    Untangler
    Join Date
    Jun 2008
    Posts
    55

    Default

    Quote Originally Posted by sky-knight View Post
    I agree, that doesn't seem stressed at all, which either means your VM is corrupt or the hypervisor is really screwing up disk scheduling for some reason. That is assuming some not so obvious configuration fault is to blame.

    I'd give UT support some more time, Collin is a great guy but he's far from all knowing! And now you've gained the attention of the all mighty DMorris... Which as much as I know it pains him for me to say this, but if he make progress on this no one can!
    Agreed. I'll see if the rebuild works.

  8. #18
    Untangler
    Join Date
    Jun 2008
    Posts
    55

    Default

    Quote Originally Posted by soccerextreme View Post
    Yes, that is the correct ticket. The 4.4.1 timeout message is generated by the sender's server not the recipient server.
    It's also important to note that the 4.4.1 errors don't always accompany the outages either (maybe 50% of the time do we see just one), otherwise the SMTP probe just says says, "Unable to connect to the remote server"

  9. #19
    Untangler
    Join Date
    Jun 2008
    Posts
    55

    Default

    Okay, I've switched over to a new box, so we'll see what happens. I typically have outages every 1-2 hours on average so I should know pretty quickly if there's any change for the better or not.

  10. #20
    Untangler
    Join Date
    Jun 2008
    Posts
    55

    Default

    @dmorris, so if I understand you correctly, what you're saying is that the reason the bypass SMTP traffic fixes the problem is not because Untangle itself is directly causing an outage but rather in its attempt to scan incoming email it's adding enough latency to the incoming traffic that our Exchange server(s) give up on the transmissions periodically. Is that correct?

Page 2 of 3 FirstFirst 123 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