NetBSD Problem Report #51568

From hf@spg.tu-darmstadt.de  Fri Oct 21 15:17:31 2016
Return-Path: <hf@spg.tu-darmstadt.de>
Received: from mail.netbsd.org (mail.netbsd.org [199.233.217.200])
	(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
	(Client CN "mail.netbsd.org", Issuer "Postmaster NetBSD.org" (verified OK))
	by mollari.NetBSD.org (Postfix) with ESMTPS id 8B1D17A28E
	for <gnats-bugs@gnats.NetBSD.org>; Fri, 21 Oct 2016 15:17:31 +0000 (UTC)
Message-Id: <201610211517.u9LFHCtK011609@Zinnenwand.nt.e-technik.tu-darmstadt.de>
Date: Fri, 21 Oct 2016 17:17:12 +0200 (CEST)
From: Hauke Fath <hf@spg.tu-darmstadt.de>
Reply-To: Hauke Fath <hf@spg.tu-darmstadt.de>
To: gnats-bugs@NetBSD.org
Cc: Hauke Fath <hf@spg.tu-darmstadt.de>
Subject: base dhcpd dumps core in a CARP setup 
X-Send-Pr-Version: 3.95

>Number:         51568
>Category:       bin
>Synopsis:       base dhcpd dumps core in a CARP setup
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    bin-bug-people
>State:          closed
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri Oct 21 15:20:00 +0000 2016
>Closed-Date:    Wed Jul 21 08:04:21 +0000 2021
>Last-Modified:  Thu Jul 22 08:00:02 +0000 2021
>Originator:     Hauke Fath
>Release:        NetBSD 7.0_STABLE
>Organization:
Technische Universitaet Darmstadt
>Environment:


System: NetBSD Zinnenwand 7.0_STABLE NetBSD 7.0_STABLE (FIFI-$Revision: 1.85 $) #1: Tue Oct 18 11:19:47 CEST 2016 hf@Hochstuhl:/var/obj/netbsd-builds/7/amd64/sys/arch/amd64/compile/FIFI amd64
Architecture: x86_64
Machine: amd64
>Description:

	On a pair of machines set up as redundant router with carp(4),
	the base dhcpd(8) [v4.3.0 on netbsd-7, configured for
	failover] frequently dumps core, probably in an attempt to
	sync leases, which according to the logs does not
	happen. Beyond a

Oct 19 13:22:01 Zinnenwand /netbsd: pid 467 (dhcpd), uid 0: exited on signal 6 (core dumped)

	there is nothing in the logs; 'gdb dhcpd dhcpd.core' shows an
	assertion failure.

	Unfortunately, make(1)ing binaries from the source tree is
	severely underdocumented compared to build.sh. I wanted to
	build dhcpd with -g3; went to build named's libisc, and punted
	when that wanted to build libthread, instead of using the
	installed library.

	A debug build of the pkgsrc net/isc-dhcpd turned out lots
	easier, as well as unnecessary: its v4.3.3 works just fine.


>How-To-Repeat:

	Set up a carp(4) based redundant router pair, run dhcpd(8) on
	it with failover configured. Watch the daemon dump core on
	both sides.


>Fix:

	Pull up -current dhcp 4.3.3 to (at least) netbsd-7, I guess.

	I can test this, if anybody can provide me with a reasonable
	way of building dhcpd from -current sources against a netbsd-7
	installation...




>Release-Note:

>Audit-Trail:

From: "Hauke Fath" <hauke@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: PR/51568 CVS commit: src/etc
Date: Mon, 16 Jan 2017 09:39:26 +0000

 Module Name:	src
 Committed By:	hauke
 Date:		Mon Jan 16 09:39:26 UTC 2017

 Modified Files:
 	src/etc: protocols

 Log Message:
 Add carp as an alias for vrrp - after all, we do not ship vrrp, but we
 do ship carp(4).

 Restore the pfsync entry that was added with 1.20, then wiped out by
 the 1.21 import. Please merge any wholesale imports properly.

 Remove http://www.sethwklein.net/projects/iana-etc/ which 404s.

 Should fix PR bin/51568


 To generate a diff of this commit:
 cvs rdiff -u -r1.28 -r1.29 src/etc/protocols

 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.

State-Changed-From-To: open->feedback
State-Changed-By: dholland@NetBSD.org
State-Changed-When: Wed, 21 Jul 2021 04:46:04 +0000
State-Changed-Why:
did your commit actually fix it?


State-Changed-From-To: feedback->closed
State-Changed-By: hauke@NetBSD.org
State-Changed-When: Wed, 21 Jul 2021 08:04:21 +0000
State-Changed-Why:
Setup is long gone.


From: Hauke Fath <hf@spg.tu-darmstadt.de>
To: gnats-bugs@netbsd.org
Cc: gnats-admin@netbsd.org, dholland@NetBSD.org
Subject: Re: bin/51568 (base dhcpd dumps core in a CARP setup)
Date: Wed, 21 Jul 2021 10:02:48 +0200

 On Wed, 21 Jul 2021 04:46:04 +0000 (UTC), dholland@NetBSD.org wrote:
 > 51568
 >=20
 > did your commit actually fix it?

 Good question. It's possible, but then the whole router setup was=20
 completely unworkable; it's been migrated long ago for good.=20

 I doubt anybody uses carp on NetBSD at all.=20

 I'll close the PR.

 Cheerio,
 Hauke

 --=20
      The ASCII Ribbon Campaign                    Hauke Fath
 ()     No HTML/RTF in email            Institut f=FCr Nachrichtentechnik
 /\     No Word docs in email                     TU Darmstadt
      Respect for open standards              Ruf +49-6151-16-21344

From: Frank Kardel <kardel@netbsd.org>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: bin/51568 (base dhcpd dumps core in a CARP setup)
Date: Wed, 21 Jul 2021 14:21:36 +0200

 Well, I do have some carp setups - works for IPv4 only, is likely to 
 lock up when configuring IPv4 and IPv6 on the carp device.

 The lockup still needs to be investigated.

 As for the dhcpd aborting. I have seen that also. It had nothing to do 
 with carp other that when you

 start to use carp you are likely to use dhcpd fail over. I moved to 
 pkgsrc-isc_dhcpd and that worked.

 Thus at that time dhcpd in the base had a problem. the pkgsrc version 
 (same release) was fine.

 I haven't checked todays state of affairs though.

 Frank


 On 07/21/21 10:05, Hauke Fath wrote:
 > The following reply was made to PR bin/51568; it has been noted by GNATS.
 >
 > From: Hauke Fath <hf@spg.tu-darmstadt.de>
 > To: gnats-bugs@netbsd.org
 > Cc: gnats-admin@netbsd.org, dholland@NetBSD.org
 > Subject: Re: bin/51568 (base dhcpd dumps core in a CARP setup)
 > Date: Wed, 21 Jul 2021 10:02:48 +0200
 >
 >   On Wed, 21 Jul 2021 04:46:04 +0000 (UTC), dholland@NetBSD.org wrote:
 >   > 51568
 >   >=20
 >   > did your commit actually fix it?
 >   
 >   Good question. It's possible, but then the whole router setup was=20
 >   completely unworkable; it's been migrated long ago for good.=20
 >   
 >   I doubt anybody uses carp on NetBSD at all.=20
 >   
 >   I'll close the PR.
 >   
 >   Cheerio,
 >   Hauke
 >   
 >   --=20
 >        The ASCII Ribbon Campaign                    Hauke Fath
 >   ()     No HTML/RTF in email            Institut f=FCr Nachrichtentechnik
 >   /\     No Word docs in email                     TU Darmstadt
 >        Respect for open standards              Ruf +49-6151-16-21344
 >   

From: David Holland <dholland-bugs@netbsd.org>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: bin/51568 (base dhcpd dumps core in a CARP setup)
Date: Thu, 22 Jul 2021 07:00:57 +0000

 On Wed, Jul 21, 2021 at 12:25:01PM +0000, Frank Kardel wrote:
  >  Well, I do have some carp setups - works for IPv4 only, is likely to 
  >  lock up when configuring IPv4 and IPv6 on the carp device.
  >  
  >  The lockup still needs to be investigated.

 Is there a PR for that? If not, please file one :-)

  >  As for the dhcpd aborting. I have seen that also. It had nothing to do 
  >  with carp other that when you
  >  
  >  start to use carp you are likely to use dhcpd fail over. I moved to 
  >  pkgsrc-isc_dhcpd and that worked.
  >  
  >  Thus at that time dhcpd in the base had a problem. the pkgsrc version 
  >  (same release) was fine.
  >  
  >  I haven't checked todays state of affairs though.

 Does this warrant reopening this PR or would it be better to generate
 a new one?

 -- 
 David A. Holland
 dholland@netbsd.org

From: Frank Kardel <kardel@netbsd.org>
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: bin/51568 (base dhcpd dumps core in a CARP setup)
Date: Thu, 22 Jul 2021 09:56:01 +0200

 On 07/22/21 09:05, David Holland wrote:
 >   On Wed, Jul 21, 2021 at 12:25:01PM +0000, Frank Kardel wrote:
 >    >  Well, I do have some carp setups - works for IPv4 only, is likely to
 >    >  lock up when configuring IPv4 and IPv6 on the carp device.
 >    >
 >    >  The lockup still needs to be investigated.
 >   
 >   Is there a PR for that? If not, please file one :-)
 Sure, once I am able to collect some information about it instead of 
 reciting my memory from a year ago.
 The pattern is like this:
      A and B run.
      A is master in 4 crap instances.
      B ist backup on those instances
      A goes down
      B becomes master on those 4 instances
      A comes up and gets stuck forever in "waiting for DAD to complete"
 Unfortunatly this happens on our gateway routers to testing time is 
 limited. Maybe
 I can recreate that elsewhere but not on short notice.
 >   
 >    >  As for the dhcpd aborting. I have seen that also. It had nothing to do
 >    >  with carp other that when you
 >    >
 >    >  start to use carp you are likely to use dhcpd fail over. I moved to
 >    >  pkgsrc-isc_dhcpd and that worked.
 >    >
 >    >  Thus at that time dhcpd in the base had a problem. the pkgsrc version
 >    >  (same release) was fine.
 >    >
 >    >  I haven't checked todays state of affairs though.
 >   
 >   Does this warrant reopening this PR or would it be better to generate
 >   a new one?
 Probably a new one to get rid of the carp association.
 >   
 >   --
 >   David A. Holland
 >   dholland@netbsd.org
 >   
 Frank

>Unformatted:

Home	PR Database Search