28 Led
2020
28 Led
'20
19:58
Hello,
I have a problem with a new, small IPv6 reverse zone
It's a small zone, configured like this :
zone:
- domain: "a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa."
file: "a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa"
notify: "corrin"
acl: "acl_corrin"
dnssec-signing: "off"
module: mod-synthrecord/revovh2
There is data in the journal :
root@arrakeen:/var/lib/knot/external# knotc zone-read
a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa
[a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.]
a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa. 2560 SOA ns.geekwu.org.
hostmaster.a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa. 1580234956 16384
2048 1048576 2560
[a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.]
a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa. 259200 NS ns.geekwu.org.
[a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.]
a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa. 259200 NS ns4.geekwu.org.
[a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.]
0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.
3600 PTR imperium.geekwu.org.
[a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.]
2.2.1.0.0.0.0.0.0.0.0.0.1.0.0.0.a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.
3600 PTR git.geekwu.org.
[a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.]
a.2.a.2.0.0.0.0.0.0.0.0.1.0.0.0.a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.
3600 PTR corrin.geekwu.org.
But if I change anything (like adding another host), on knotc reload I
get this error :
knotd[21892]: error: [a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.] zone
event 'load' failed (malformed data)
I've purged the zone with zone-purge, replaced the zone file with a copy
of the dump, and got a new record on reload, but further edits fails the
same way
I'm using knot 2.9.2-1~cz.nic~buster1 for deb.knot-dns.cz/knot-latest
Do you have an idea on what's going on ?
Thanks,
--
Bastien Durel
28 Led
28 Led
20:22
Hello Bastien,
Could you share the zone file and its exact modification?
Daniel
On 2020-01-28 19:58, Bastien Durel wrote:
Hello,
I have a problem with a new, small IPv6 reverse zone
It's a small zone, configured like this :
zone:
- domain: "a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa."
file: "a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa"
notify: "corrin"
acl: "acl_corrin"
dnssec-signing: "off"
module: mod-synthrecord/revovh2
There is data in the journal :
root@arrakeen:/var/lib/knot/external# knotc zone-read
a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa
[a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.]
a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa. 2560 SOA ns.geekwu.org.
hostmaster.a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa. 1580234956 16384
2048 1048576 2560
[a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.]
a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa. 259200 NS ns.geekwu.org.
[a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.]
a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa. 259200 NS ns4.geekwu.org.
[a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.]
0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.
3600 PTR imperium.geekwu.org.
[a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.]
2.2.1.0.0.0.0.0.0.0.0.0.1.0.0.0.a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.
3600 PTR git.geekwu.org.
[a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.]
a.2.a.2.0.0.0.0.0.0.0.0.1.0.0.0.a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.
3600 PTR corrin.geekwu.org.
But if I change anything (like adding another host), on knotc reload I
get this error :
knotd[21892]: error: [a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.] zone
event 'load' failed (malformed data)
I've purged the zone with zone-purge, replaced the zone file with a
copy of the dump, and got a new record on reload, but further edits
fails the same way
I'm using knot 2.9.2-1~cz.nic~buster1 for deb.knot-dns.cz/knot-latest
Do you have an idea on what's going on ?
Thanks,
--
Bastien Durel
29 Led
29 Led
14:04
Hello Bastien,
Could you share the zone file and its exact modification?
Daniel
Hello.
(Your mail was lost in migration, so it will probably break the thread.)
I've attached zone.dump.txt, which is the zone file that were dumped by
knot after a shutdown; and zone.bad.txt, which was the modified file
I've edited and knot refuses to load
Regards,
--
Bastien Durel
14:10
On 2020-01-29 14:04, Bastien Durel wrote:
I've re-edited the zone file right now, with aven more minor change, and
get the same error as before : janv. 29 14:07:28 arrakeen knotd[7971]:
error: [a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.] zone event 'load'
failed (malformed data)
--
Bastien Durel
Hello
Bastien,
Could you share the zone file and its exact modification?
Daniel
Hello.
(Your mail was lost in migration, so it will probably break the thread.)
I've attached zone.dump.txt, which is the zone file that were dumped by
knot after a shutdown; and zone.bad.txt, which was the modified file
I've edited and knot refuses to load
Regards,
14:14
Hi Bastien,
at my place, all your zone files tend to load normally.
Maybe the "malformed" something comes from another place than a zone file.
Could you please share some longer snippet of your logs?
Have you tried to purge the journal (if it makes no harm in your use-case)?
Libor
Dne 29.01.20 v 14:10 Bastien Durel napsal(a):
On 2020-01-29 14:04, Bastien Durel wrote:
I've re-edited the zone file right now, with aven more minor change,
and get the same error as before : janv. 29 14:07:28 arrakeen
knotd[7971]: error: [a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa.] zone
event 'load' failed (malformed data)
Hello
Bastien,
Could you share the zone file and its exact modification?
Daniel
Hello.
(Your mail was lost in migration, so it will probably break the thread.)
I've attached zone.dump.txt, which is the zone file that were dumped
by knot after a shutdown; and zone.bad.txt, which was the modified
file I've edited and knot refuses to load
Regards,
14:32
Hi Bastien,
at my place, all your zone files tend to load normally.
Maybe the "malformed" something comes from another place than a zone file.
Could you please share some longer snippet of your logs?
Have you tried to purge the journal (if it makes no harm in your use-case)?
Libor
Hello,
Here is the complete log of the reload event.
(I do not get messages from the list, was my address blacklisted due to
errors, or is the delivery failing due to old records in DNS ?)
I purged the journal with
"knotc zone-purge a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa", is it a
better/stronger way ?
Thanks,
--
Bastien Durel
14:32
Hi Bastien,
at my place, all your zone files tend to load normally.
Maybe the "malformed" something comes from another place than a zone file.
Could you please share some longer snippet of your logs?
Have you tried to purge the journal (if it makes no harm in your use-case)?
Libor
Hello,
Here is the complete log of the reload event.
(I do not get messages from the list, was my address blacklisted due to
errors, or is the delivery failing due to old records in DNS ?)
I purged the journal with
"knotc zone-purge a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa", is it a
better/stronger way ?
Thanks,
--
Bastien Durel
15:31
Bastien,
The zone purge operation requires the force mode:
"knotc -f zone-purge a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa"
Does this help?
Daniel
On 1/29/20 2:32 PM, Bastien Durel wrote:
Hi Bastien,
at my place, all your zone files tend to load normally.
Maybe the "malformed" something comes from another place than a zone file.
Could you please share some longer snippet of your logs?
Have you tried to purge the journal (if it makes no harm in your use-case)?
Libor
Hello,
Here is the complete log of the reload event.
(I do not get messages from the list, was my address blacklisted due to errors, or is the
delivery failing due to old records in DNS ?)
I purged the journal with
"knotc zone-purge a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa", is it a
better/stronger way ?
Thanks,
15:36
Bastien,
The zone purge operation requires the force mode:
"knotc -f zone-purge a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa"
Does this help?
Daniel
I indeed used the -f flag, as instructed by knotc.
This allowed me to add the last 2 entries in the zone, but the problem
reappeared soon after that.
--
Bastien
15:47
Ok,
Then try removing the journal directory.
Btw, is the same after server restart?
On 1/29/20 3:36 PM, Bastien Durel wrote:
Bastien,
The zone purge operation requires the force mode:
"knotc -f zone-purge a.8.1.a.8.0.0.0.0.d.1.4.1.0.0.2.ip6.arpa"
Does this help?
Daniel
I indeed used the -f flag, as instructed by knotc.
This allowed me to add the last 2 entries in the zone, but the problem
reappeared soon after that.
16:06
Ok,
Then try removing the journal directory.
Btw, is the same after server restart?
Removing the journal directory allowed me to make a bunch of changes.
The server was restarted a few times during the problem. That's why I
got my zone file overwritten by the dump ^^
I kept the journal directory in my /tmp, if you want to look at it. If
not, no problem ;)
Thanks,
--
Bastien Durel
16:08
If you could send us a tarball including journal, config file, and zone file it would be
nice :-)
On 1/29/20 4:06 PM, Bastien Durel wrote:
Ok,
Then try removing the journal directory.
Btw, is the same after server restart?
Removing the journal directory allowed me to make a bunch of changes.
The server was restarted a few times during the problem. That's why I
got my zone file overwritten by the dump ^^
I kept the journal directory in my /tmp, if you want to look at it. If
not, no problem ;)
Thanks,
19:14
On 2020-01-29 16:08, Daniel Salzman wrote:
If you could send us a tarball including journal,
config file, and zone file it would be nice :-)
On 1/29/20 4:06 PM, Bastien Durel wrote:
>> Ok,
>>
>> Then try removing the journal directory.
>>
>> Btw, is the same after server restart?
>
> Removing the journal directory allowed me to make a bunch of changes.
> The server was restarted a few times during the problem. That's why I
> got my zone file overwritten by the dump ^^
>
> I kept the journal directory in my /tmp, if you want to look at it. If
> not, no problem ;)
>
> Thanks,
>
Sent it off-list.
Regards,
--
Bastien Durel
30 Led
30 Led
16:46
Hi Bastien,
thank you very much for all the reports, information and data.
I hope i have found and fixed the issue.
The issue can be workarounded by deleting the whole journal (rm -rf
/path/to/journal/), if this makes no collateral damage.
It will be fixed by this patch:
https://gitlab.labs.nic.cz/knot/knot-dns/merge_requests/1104
You might check and verify my conclusions.
Thanks,
Libor
Dne 29.01.20 v 19:14 Bastien Durel napsal(a):
On 2020-01-29 16:08, Daniel Salzman wrote:
If you could send us a tarball including journal,
config file, and
zone file it would be nice :-)
On 1/29/20 4:06 PM, Bastien Durel wrote:
>> Ok,
>>
>> Then try removing the journal directory.
>>
>> Btw, is the same after server restart?
>
> Removing the journal directory allowed me to make a bunch of changes.
> The server was restarted a few times during the problem. That's why I
> got my zone file overwritten by the dump ^^
>
> I kept the journal directory in my /tmp, if you want to look at it. If
> not, no problem ;)
>
> Thanks,
>
Sent it off-list.
Regards,
17:25
Le jeudi 30 janvier 2020 à 16:46 +0100, libor.peltan a écrit :
Hi Bastien,
thank you very much for all the reports, information and data.
I hope i have found and fixed the issue.
The issue can be workarounded by deleting the whole journal (rm -rf
/path/to/journal/), if this makes no collateral damage.
It will be fixed by this patch:
https://gitlab.labs.nic.cz/knot/knot-dns/merge_requests/1104
You might check and verify my conclusions.
Thanks,
Libor
Hello.
I deleted my journal yesterday, so I didn't have a test case anymore ;)
Thanks for your work.
--
Bastien
1760
days inactive
1762
days old
14 comments
5 participants
participants (5)
-
Bastien Durel -
Bastien Durel -
Daniel Salzman -
daniel.salzman＠nic.cz
-
libor.peltan