Bug 30580
| Summary: | named crashes on 4.4-PRERELEASE | ||
|---|---|---|---|
| Product: | Base System | Reporter: | ask |
| Component: | i386 | Assignee: | freebsd-bugs (Nobody) <bugs> |
| Status: | Closed FIXED | ||
| Severity: | Affects Only Me | ||
| Priority: | Normal | ||
| Version: | Unspecified | ||
| Hardware: | Any | ||
| OS: | Any | ||
On Fri, Sep 14, 2001 at 05:21:24PM -0700, Ask Bjoern Hansen wrote:
> It's both running as a (somewhat busy) authoritative server and as a (very very low traffic) resolving server.
> >How-To-Repeat:
> I start the named and wait a few hours or half a day and I'll have a new core file. :-)
This is usually due to problems with your system clock.
Kris
On Fri, Sep 14, 2001 at 05:21:24PM -0700, Ask Bjoern Hansen wrote:
> It's both running as a (somewhat busy) authoritative server and as a (very very low traffic) resolving server.
> >How-To-Repeat:
> I start the named and wait a few hours or half a day and I'll have a new core file. :-)
This is usually due to problems with your system clock.
Kris
On Fri, Sep 14, 2001 at 05:21:24PM -0700, Ask Bjoern Hansen wrote:
> It's both running as a (somewhat busy) authoritative server and as a (very very low traffic) resolving server.
> >How-To-Repeat:
> I start the named and wait a few hours or half a day and I'll have a new core file. :-)
This is usually due to problems with your system clock.
Kris
On Sat, 15 Sep 2001, Kris Kennaway wrote: > On Fri, Sep 14, 2001 at 05:21:24PM -0700, Ask Bjoern Hansen wrote: > > > It's both running as a (somewhat busy) authoritative server and as a (very very low traffic) resolving server. > > >How-To-Repeat: > > I start the named and wait a few hours or half a day and I'll have a new core file. :-) > > This is usually due to problems with your system clock. As in the clock keeping track of the time of day? That sounds really odd. I searched the mail archives for more references about this, but didn't find anything. Can you point me in the right direction? I've recently had problems with the IDE (don't ask) drive and haven't had time to replace it yet; could that be related? - ask -- ask bjoern hansen, http://ask.netcetera.dk/ !try; do(); On Sat, Sep 15, 2001 at 05:32:17AM -0700, Ask Bjoern Hansen wrote: > On Sat, 15 Sep 2001, Kris Kennaway wrote: > > > On Fri, Sep 14, 2001 at 05:21:24PM -0700, Ask Bjoern Hansen wrote: > > > > > It's both running as a (somewhat busy) authoritative server and as a (very very low traffic) resolving server. > > > >How-To-Repeat: > > > I start the named and wait a few hours or half a day and I'll have a new core file. :-) > > > > This is usually due to problems with your system clock. > > As in the clock keeping track of the time of day? That sounds really > odd. I searched the mail archives for more references about this, > but didn't find anything. Can you point me in the right direction? Search the -security mailing list for "named assertion failed". Since you didn't post the actual error message it's hard to verify it's the same thing. > I've recently had problems with the IDE (don't ask) drive and > haven't had time to replace it yet; could that be related? I haven't heard of that, but it's not impossible. Kris On Sat, Sep 15, 2001 at 01:50:02PM -0700, Kris Kennaway wrote:
> On Sat, Sep 15, 2001 at 05:32:17AM -0700, Ask Bjoern Hansen wrote:
> > On Sat, 15 Sep 2001, Kris Kennaway wrote:
> >=20
> > > On Fri, Sep 14, 2001 at 05:21:24PM -0700, Ask Bjoern Hansen wrote:
> > >
> > > > It's both running as a (somewhat busy) authoritative server and as a =
> (very very low traffic) resolving server.
> > > > >How-To-Repeat:
> > > > I start the named and wait a few hours or half a day and I'll have a =
> new core file. :-)
> > >
> > > This is usually due to problems with your system clock.
> >=20
> > As in the clock keeping track of the time of day? That sounds really
> > odd. I searched the mail archives for more references about this,
> > but didn't find anything. Can you point me in the right direction?
>
> Search the -security mailing list for "named assertion failed". Since
> you didn't post the actual error message it's hard to verify it's the
> same thing.
>
> > I've recently had problems with the IDE (don't ask) drive and
> > haven't had time to replace it yet; could that be related?
>
> I haven't heard of that, but it's not impossible.
Did you follow this up yet?
Kris
On Sun, 23 Sep 2001, Kris Kennaway wrote: > > >=20 > > > > On Fri, Sep 14, 2001 at 05:21:24PM -0700, Ask Bjoern Hansen wrote: > > > > > > > > > It's both running as a (somewhat busy) authoritative server and as a = > > (very very low traffic) resolving server. > > > > > >How-To-Repeat: > > > > > I start the named and wait a few hours or half a day and I'll have a = > > new core file. :-) > > > > > > > > This is usually due to problems with your system clock. > > >=20 > > > As in the clock keeping track of the time of day? That sounds really > > > odd. I searched the mail archives for more references about this, > > > but didn't find anything. Can you point me in the right direction? > > > > Search the -security mailing list for "named assertion failed". Since > > you didn't post the actual error message it's hard to verify it's the > > same thing. > > > > > I've recently had problems with the IDE (don't ask) drive and > > > haven't had time to replace it yet; could that be related? > > > > I haven't heard of that, but it's not impossible. > > Did you follow this up yet? no; I was actually just going to do so to tell that it seems resolved. I replaced the faulty drive and it had not crashed since (and actually it took a break in the last few days before I replaced it). Anyway, it crashed again: (gdb) bt #0 0x28118764 in kill () from /usr/lib/libc.so.4 #1 0x281547b6 in abort () from /usr/lib/libc.so.4 #2 0x807456d in ns_panic (category=18, dump_core=1, format=0x80b7e62 "%s:%d: %s(%s)%s%s failed.") at /usr/src/usr.sbin/named/../../contrib/bind/bin/named/ns_glue.c:176 #3 0x80745e7 in ns_assertion_failed ( file=0x80b1da0 "/usr/src/usr.sbin/named/../../contrib/bind/bin/named/ns_main.c", line=552, type=assert_insist, cond=0x80b1f3b "errno == EINTR", print_errno=1) at /usr/src/usr.sbin/named/../../contrib/bind/bin/named/ns_glue.c:185 #4 0x805fcc3 in main (argc=1, argv=0xbfbffde0, envp=0xbfbffde4) at /usr/src/usr.sbin/named/../../contrib/bind/bin/named/ns_main.c:552 from syslog: Sep 23 06:59:12 miette named[201]: /usr/src/usr.sbin/named/../../contrib/bind/bin/named/ns_main. c:552: INSIST(errno == EINTR): Invalid argument failed. Sep 23 06:59:12 miette named[201]: /usr/src/usr.sbin/named/../../contrib/bind/bin/named/ns_main. c:552: INSIST(errno == EINTR): Invalid argument failed. Sep 23 06:59:14 miette /kernel: pid 201 (named), uid 0: exited on signal 6 (core dumped) There are no other messages around the same time. No funny hardware issues and the box is using NTP and I haven't seen any messages from ntpd about funny things going on with the time on the box. (Which was what the archived messages about the named assertion failure indicated). :-/ Any clues would be most appreciated. - ask -- ask bjoern hansen, http://ask.netcetera.dk/ !try; do(); On Mon, Sep 24, 2001 at 03:56:14AM -0700, Ask Bjoern Hansen wrote:
> from syslog:
>
> Sep 23 06:59:12 miette named[201]: /usr/src/usr.sbin/named/../../contrib/bind/bin/named/ns_main. c:552: INSIST(errno == EINTR): Invalid argument failed.
> Sep 23 06:59:12 miette named[201]: /usr/src/usr.sbin/named/../../contrib/bind/bin/named/ns_main. c:552: INSIST(errno == EINTR): Invalid argument failed.
> Sep 23 06:59:14 miette /kernel: pid 201 (named), uid 0: exited on signal 6 (core dumped)
>
> There are no other messages around the same time. No funny hardware
> issues and the box is using NTP and I haven't seen any messages from
> ntpd about funny things going on with the time on the box. (Which was
> what the archived messages about the named assertion failure
> indicated). :-/
>
> Any clues would be most appreciated.
You probably should take this up on the BIND support list.
Kris
State Changed From-To: open->feedback Did you try the BIND in the latest STABLE? State Changed From-To: feedback->closed Feedback timeout |
my named crashes every few hours with this, (gdb) bt #0 0x28118764 in kill () from /usr/lib/libc.so.4 #1 0x281547b6 in abort () from /usr/lib/libc.so.4 #2 0x807456d in ns_panic (category=18, dump_core=1, format=0x80b7e62 "%s:%d: %s(%s)%s%s failed.") at /usr/src/usr.sbin/named/../../contrib/bind/bin/named/ns_glue.c:176 #3 0x80745e7 in ns_assertion_failed ( file=0x80b1da0 "/usr/src/usr.sbin/named/../../contrib/bind/bin/named/ns_main.c", line=552, type=assert_insist, cond=0x80b1f3b "errno == EINTR", print_errno=1) at /usr/src/usr.sbin/named/../../contrib/bind/bin/named/ns_glue.c:185 #4 0x805fcc3 in main (argc=1, argv=0xbfbffde0, envp=0xbfbffde4) at /usr/src/usr.sbin/named/../../contrib/bind/bin/named/ns_main.c:552 It's both running as a (somewhat busy) authoritative server and as a (very very low traffic) resolving server. How-To-Repeat: I start the named and wait a few hours or half a day and I'll have a new core file. :-)