DaemonForums  

Go Back   DaemonForums > OpenBSD > OpenBSD General

OpenBSD General Other questions regarding OpenBSD which do not fit in any of the categories below.

Reply
 
Thread Tools Display Modes
  #1   (View Single Post)  
Old 14th July 2017
hanzer's Avatar
hanzer hanzer is offline
Real Name: Adam Jensen
just passing through
 
Join Date: Oct 2013
Location: EST USA
Posts: 234
Default too much network activity kills the interface

6.0 GENERIC#1 i386

$ dmesg | nc termbin.com 9999
http://termbin.com/yvz7

This machine (above) is used as the Internet gateway for my LAN. If too much Internet activity of a certain type is generated, the internal network interface (re0) will shut down. Example: a LAN machine runs a torrent client with 30+ active downloads.

I am a little surprised and frustrated by this glass jaw behavior but in retrospect, I think it might be a pervasive theme: OpenBSD doesn't seem to incorporate a fault-tolerant and fail-safe design paradigm. Side note: Besides the re0 interface occasionally collapsing, and attack from the Tor network (I assume it is an attack) can cause the entire system to lock. Another example, I had OpenBSD on a laptop and simply turning on the wireless network card (for which there was a driver and the card type was presumably supposed to work) would consistently cause the laptop to completely freeze.

Anyway, is there any network tuning advice or logging+diagnostic advice that might bolster this fainting little pansy of a computer system? <ouch!>

I am really beginning to see the OpenBSD philosophy as being little more than a reflection of the tyranny of a petty dictator mentality that will shit itself and pass-out at the first sign of trouble. (too polemic to tolerate?)
Reply With Quote
  #2   (View Single Post)  
Old 14th July 2017
jggimi's Avatar
jggimi jggimi is offline
More noise than signal
 
Join Date: May 2008
Location: USA
Posts: 5,891
Default

  1. There's been a lot of network stack development since 6.0 was tagged for release in July of 2016.
  2. There have been 5 or 6 RELIABILITY errata patches posted for 6.0 since you built this -stable kernel, including fixes to PF for integer overflows and kernel memory leaks with ICMP packets.
I don't know if updating your -stable system will eliminate the problems you've experienced. As well, the problems you are experiencing could have a hardware-related root cause. But either way, it's clear to me this system is not well-maintained, as it does not include the most recent published patches. Nor is it the most recent release.

If you wished to keep using the OS, you might consider installing a -current test system on a USB disk, and testing with Tor and Torrent, and see if you can recreate the same problems.

Two additional comments:

  • This forum is not an official support channel. We're just users trying to help each other. If you want to report a bug, the bugs@ mailing list is the appropriate place. But if you do, expect to be instructed to maintain your 6.0 system, since it is not up-to-date.
  • Since you are unhappy with the OS and with the Project, consider switching to an OS that will make you happier. You won't have to put up with these problems, perhaps. Certainly not with those people. Or with me and my snarky replies (such as this one).

Last edited by jggimi; 14th July 2017 at 11:14 PM. Reason: typos
Reply With Quote
  #3   (View Single Post)  
Old 15th July 2017
hanzer's Avatar
hanzer hanzer is offline
Real Name: Adam Jensen
just passing through
 
Join Date: Oct 2013
Location: EST USA
Posts: 234
Default

Quote:
Originally Posted by jggimi View Post
  1. There's been a lot of network stack development since 6.0 was tagged for release in July of 2016.
  2. There have been 5 or 6 RELIABILITY errata patches posted for 6.0 since you built this -stable kernel, including fixes to PF for integer overflows and kernel memory leaks with ICMP packets.
Interesting, thanks. So it's not a matter of examining some log files and twiddling some values in a configuration file...

Quote:
Originally Posted by jggimi View Post
I don't know if updating your -stable system will eliminate the problems you've experienced. As well, the problems you are experiencing could have a hardware-related root cause. But either way, it's clear to me this system is not well-maintained, as it does not include the most recent published patches. Nor is it the most recent release.
"not well-maintained"? Ouch!

I kind of expected basic network functionality to work well by now. Live and learn...

Quote:
Originally Posted by jggimi View Post
  • This forum is not an official support channel. We're just users trying to help each other. If you want to report a bug, the bugs@ mailing list is the appropriate place. But if you do, expect to be instructed to maintain your 6.0 system, since it is not up-to-date.
  • Since you are unhappy with the OS and with the Project, consider switching to an OS that will make you happier. You won't have to put up with these problems, perhaps. Certainly not with those people. Or with me and my snarky replies (such as this one).
Sometimes progress requires shaking things up a bit. I heard somewhere that there are three basic categories of people: creators of culture, maintainers of culture, and destroyers of culture - (oh gee, was that a Hitler quote?) - anyway, the point is that while it's reasonable to appreciate those who maintain the status quo, it also seems reasonable to avoid stagnation and maybe even make progress. Lectures are fun, aren't they?

As always, thanks for your expertise and stellar technical communication. Without you here, I probably would have retired the OpenBSD hobby long ago.
Reply With Quote
  #4   (View Single Post)  
Old 15th July 2017
jggimi's Avatar
jggimi jggimi is offline
More noise than signal
 
Join Date: May 2008
Location: USA
Posts: 5,891
Default

Quote:
Originally Posted by hanzer View Post
Interesting, thanks. So it's not a matter of examining some log files and twiddling some values in a configuration file...
I don't know if there is a log file that would show anything when your interface "shuts down" but you might find a report in /var/log/messages, which is a log of the system console. But when your system enters the "completely freeze" state, I doubt anything is logged during those occurrences. There is the kernel debugger ddb(4), but that may or may not be able to function during the event. Even if it does, would you have the interest in using it to debug the problem?
Quote:
I kind of expected basic network functionality to work well by now.
It's been working fine for me for the 13 years I've been using the OS. Have I had problems with OpenBSD systems? Of course. Have there been software problems? Yes. Have there been hardware problems? Yes.

From the information posted, I can't tell you why you're having the problem/problems. It could be the OS. It could be something you've provisioned. It could be your hardware -- even a network cable could be the root cause.

If you switched to an OS that you dislike less, would it suck less for you? You'll have to make the switch to know for sure.

Last edited by jggimi; 15th July 2017 at 04:42 AM. Reason: there may be multiple problems
Reply With Quote
  #5   (View Single Post)  
Old 15th July 2017
hanzer's Avatar
hanzer hanzer is offline
Real Name: Adam Jensen
just passing through
 
Join Date: Oct 2013
Location: EST USA
Posts: 234
Default

Quote:
Originally Posted by jggimi View Post
There is the kernel debugger ddb(4), but that may or may not be able to function during the event. Even if it does, would you have the interest in using it to debug the problem?
Under the circumstances, wouldn't that be a huge waste of time and effort?

Quote:
Originally Posted by jggimi View Post
If you switched to an OS that you dislike less, would it suck less for you?
I use Scientific Linux on a laptop, workstation, and a server; a media center runs Ubuntu/Kodi; the headless little Internet gateway machine runs OpenBSD. Based on the info in your first post, I'll probably cvsup to the most recent 6.0-stable and build then just live with it until the machine is ready for a trip to the Goodwill.

Quote:
Originally Posted by jggimi View Post
You'll have to make the switch to know for sure.
Are you handing me my hat? Subtle

+1 for the Maintainers, but it's a sad day for the designers. <dramatic smirk>

Realistically, I'm probably too exhausted today for a critical analysis or speculative discussion of architectural principles and paradigms anyway. I'm just bored. Thanks for the chat, jiggii.
Reply With Quote
  #6   (View Single Post)  
Old 15th July 2017
jggimi's Avatar
jggimi jggimi is offline
More noise than signal
 
Join Date: May 2008
Location: USA
Posts: 5,891
Default

Quote:
Originally Posted by hanzer View Post
Under the circumstances, wouldn't that be a huge waste of time and effort?
You clearly don't want to invest any significant time into this system - and it is likely you would need to develop additional technical skills before you would be able to successfully use ddb() to analyze the problem, even if it happened to be functional at the time of the next "freeze" occurrence.
Quote:
I'll probably cvsup to the most recent 6.0-stable and build then just live with it until the machine is ready for a trip to the Goodwill.
Based on my first post, if you decide not to replace OpenBSD immediately, I would first recommend seeing if the problem were repeatable with a -current test system, and if that resolves the problem/problems I'd recommend either upgrading to -current, or at least upgrading to 6.1 (due to its network stack changes) and running the new syspatch(8) tool to apply all the errata patches to-date, and add syspatch() to /etc/daily.local to keep it up-to-date.
Quote:
Are you handing me my hat?
You're the one who dislikes the OS and its Project. I cannot tell you if the problems will repeat with some other OS, but better you use an OS of your own preference to one which is anathema.
Reply With Quote
  #7   (View Single Post)  
Old 15th July 2017
hanzer's Avatar
hanzer hanzer is offline
Real Name: Adam Jensen
just passing through
 
Join Date: Oct 2013
Location: EST USA
Posts: 234
Default

Oops, that doesn't look like playful banter anymore. Sorry, I'm out.
Reply With Quote
  #8   (View Single Post)  
Old 15th July 2017
jggimi's Avatar
jggimi jggimi is offline
More noise than signal
 
Join Date: May 2008
Location: USA
Posts: 5,891
Default

I'm sorry, but I never interpreted anything in this thread to be playful banter. I assume you meant it when you stated intense dislike of the OS, its Project, its support, and its use.
Reply With Quote
  #9   (View Single Post)  
Old 8th August 2017
hanzer's Avatar
hanzer hanzer is offline
Real Name: Adam Jensen
just passing through
 
Join Date: Oct 2013
Location: EST USA
Posts: 234
Default Update

I decided to retire the old machine and replace it with a newer used machine.

$ dmesg | nc termbin.com 9999
Code:
http://termbin.com/c1dv
So far, the type of network failure that the old machine experienced hasn't recurred.
Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Slow network interface roggy OpenBSD Security 4 7th April 2017 05:17 PM
OpenBDS 5.5 Problems with network interface during installation. tastelessbit OpenBSD Installation and Upgrading 9 29th August 2014 07:47 PM
Changing Which Interface Is Assigned To Network Interface (physical) Port EverydayDiesel OpenBSD General 2 18th July 2014 10:25 AM
Device-agnostic network interface sean OpenBSD General 2 28th January 2010 06:24 PM


All times are GMT. The time now is 08:32 AM.


Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Content copyright © 2007-2010, the authors
Daemon image copyright ©1988, Marshall Kirk McKusick