[knot-dns-users] Knot on Raspberry Pi (armhf)

Ondřej Surý ondrej.sury at nic.cz
Thu Sep 17 10:26:10 CEST 2015


Hi Ulrich,<br><br>thank you for the bug report.  Could you produce a full bt after a crash?<br><br>run:<br><br>thread apply all bt full<br><br>in gdb after the crash<br><br>Also we now have official packages for armhf available from http://deb.knot-dns.cz/knot/ and if you experience crashes also with the packages please install knot-dbg before producing backtraces.<br><br>As a side note it might be more convenient to paste the long traces as part of bug report in the gitlab: https://gitlab.labs.nic.cz/labs/knot<br><br>Cheers,<br>Ondrej 

-- 
Ondřej Surý -- Technical Fellow 
-------------------------------------------- 
CZ.NIC, z.s.p.o. -- Laboratoře CZ.NIC 
Milesovska 5, 130 00 Praha 3, Czech Republic 
mailto:ondrej.sury at nic.cz https://nic.cz/ 
-------------------------------------------- 

> From: "Ulrich Wisser" <ulrich at wisser.se>
> To: knot-dns-users at lists.nic.cz
> Sent: Wednesday, September 16, 2015 10:41:37 PM
> Subject: [knot-dns-users] Knot on Raspberry Pi (armhf)

> Hello!
> I have tried to get the latest version of Knot running on my Raspberry Pi 2.
> Unfortunately does Raspian only have packages for an old version. I updated my
> Pi to debian jessie. There is actually a version of Raspian based on jessie.
> That version of Raspian does come with slightly newer packages, but still not
> up to date.
> So I compiled my own version. 2.1.0-dev
> Everything works fine until the moment I try to start knotd. IT will crash with
> a segfault.

> I did recompile with --enable-debug but I do not get any debug output in syslog.

> I did run with gdb and got the following information

> SYSLOG:

> Sep 16 22:36:57 localhost knotd[3522]: info: Knot DNS 2.1.0-dev starting

> Sep 16 22:36:57 localhost knotd[3522]: info: binding to interface
> '2001:470:28:193::53 at 53'

> Sep 16 22:36:57 localhost knotd[3522]: info: configured 1 zones

> Sep 16 22:36:57 localhost knotd[3522]: info: changing GID to '1003'

> Sep 16 22:36:57 localhost knotd[3522]: info: changing UID to '1002'

> Sep 16 22:36:57 localhost knotd[3522]: info: loading zones

> Sep 16 22:36:57 localhost knotd[3522]: info: [ example.com ] zone will be
> loaded, serial 0

> Sep 16 22:36:57 localhost knotd[3522]: info: starting server

> GDB:

> [New Thread 0x50baa200 (LWP 3526)]

> [New Thread 0x50aaa200 (LWP 3527)]

> Program received signal SIGSEGV, Segmentation fault.

> [Switching to Thread 0x50baa200 (LWP 3526)]

> scanner_process (scanner=0x50978008) at knot/zone/zonefile.c:152

> 152 if (zc->ret != KNOT_EOK) {

> (gdb) bt

> #0 scanner_process (scanner=0x50978008) at knot/zone/zonefile.c:152

> #1 0x76d80a14 in ?? () from /usr/lib/arm-linux-gnueabihf/libzscanner.so.0

> Backtrace stopped: previous frame identical to this frame (corrupt stack?)

> (gdb) l

> 147

> 148 /*! \brief Creates RR from parser input, passes it to handling function. */

> 149 static void scanner_process(zs_scanner_t *scanner)

> 150 {

> 151 zcreator_t *zc = scanner->data;

> 152 if (zc->ret != KNOT_EOK) {

> 153 scanner->stop = true;

> 154 return;

> 155 }

> 156

> (gdb)

> It seems that the scanner thread is somehow missing. But why? Anybody with
> experience with knot on armhf?

> /Ulrich
> --
> Ulrich Wisser
> ulrich at wisser.se

> _______________________________________________
> knot-dns-users mailing list
> knot-dns-users at lists.nic.cz
> https://lists.nic.cz/cgi-bin/mailman/listinfo/knot-dns-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nic.cz/pipermail/knot-dns-users/attachments/20150917/aa7d70a8/attachment.html>


More information about the knot-dns-users mailing list