6 Led
2026
6 Led
'26
4:33
在 2026/1/5 23:10, Daniel Salzman via knot-dns-users 写道:
2026-01-06T10:24:33+0800
info: Knot DNS 3.5.2 starting
2026-01-06T10:24:33+0800 info: loaded configuration file
'/home/gtld/knot/chroot4/etc/knot.conf', mapsize 500 MiB
2026-01-06T10:24:33+0800 info: using UDP reuseport, socket affinity,
incoming TCP Fast Open
2026-01-06T10:24:33+0800 info: binding to interface 10.0.0.53@53
2026-01-06T10:24:33+0800 info: binding to interface 10.0.1.53@53
2026-01-06T10:24:33+0800 info: binding to interface 10.10.3.154@53
2026-01-06T10:24:33+0800 info: binding to interface 10.10.4.154@53
2026-01-06T10:24:33+0800 info: binding to interface 10.1.136.154@53
2026-01-06T10:24:33+0800 info: binding to QUIC interface 10.10.3.154@853
2026-01-06T10:24:33+0800 info: binding to QUIC interface 10.10.4.154@853
2026-01-06T10:24:33+0800 info: binding to QUIC interface 10.1.136.154@853
2026-01-06T10:24:33+0800 info: binding to TLS interface 10.10.3.154@853
2026-01-06T10:24:33+0800 info: binding to TLS interface 10.10.4.154@853
2026-01-06T10:24:33+0800 info: binding to TLS interface 10.1.136.154@853
2026-01-06T10:24:33+0800 debug: QUIC/TLS, using self-generated key
'/home/gtld/knot/chroot4/zone/keys/quic_key.pem' with one-time certificate
2026-01-06T10:24:33+0800 info: QUIC/TLS, certificate public key
w23/TdHAJ7kwJMrb/rz/wYCDaYBVoHv4uBc880pSgCo=
2026-01-06T10:24:33+0800 info: changing GID to 1000
2026-01-06T10:24:33+0800 info: changing UID to 1000
2026-01-06T10:24:33+0800 info: process not allowed to set capabilities,
skipping
2026-01-06T10:24:33+0800 info: changed directory to /
2026-01-06T10:24:33+0800 info: loading 1 zones
2026-01-06T10:24:33+0800 debug: suspending zone events
2026-01-06T10:24:33+0800 debug: suspended zone events
2026-01-06T10:24:33+0800 info: [foo.] zone will be loaded
2026-01-06T10:24:33+0800 debug: resumed zone events
2026-01-06T10:24:33+0800 info: control, binding to
'/home/gtld/knot/chroot4/var/run/knot.sock'
2026-01-06T10:24:33+0800 info: starting server as a daemon, PID 327676
2026-01-06T10:24:37+0800 info: [foo.] zone file loaded, serial 2025123118
2026-01-06T10:24:38+0800 info: [foo.] loaded, serial none -> 2025123118,
92000117 bytes
2026-01-06T10:24:38+0800 info: [foo.] notify, outgoing, remote
10.1.136.159@53 TCP, serial 2025123118
2026-01-06T10:24:38+0800 info: [foo.] notify, outgoing, remote
10.1.136.156@53 TCP, serial 2025123118
2026-01-06T10:24:38+0800 info: server started
2026-01-06T10:24:38+0800 debug: [foo.] ACL, allowed, action transfer,
remote 10.1.136.156@40843 QUIC cert-key
kulz9ehQf5Ycn/+2mCicUdfTMuDXHbQEWBwg5qDi0Eo=
2026-01-06T10:24:38+0800 info: [foo.] IXFR, outgoing, remote
10.1.136.156@40843 QUIC, incomplete history, serial 2025123117, fallback
to AXFR
2026-01-06T10:24:38+0800 debug: [foo.] ACL, allowed, action transfer,
remote 10.1.136.156@40843 QUIC cert-key
kulz9ehQf5Ycn/+2mCicUdfTMuDXHbQEWBwg5qDi0Eo=
2026-01-06T10:24:38+0800 info: [foo.] AXFR, outgoing, remote
10.1.136.156@40843 QUIC, started, serial 2025123118
2026-01-06T10:24:39+0800 info: [foo.] AXFR, outgoing, remote
10.1.136.156@40843 QUIC, buffering finished, 0.31 seconds, 3695
messages, 62246340 bytes
2026-01-06T10:24:42+0800 debug: QUIC, terminated inactive connections 1
2026-01-06T10:24:42+0800 debug: QUIC, terminated inactive connections 1
2026-01-06T10:28:03+0800 debug: [foo.] ACL, allowed, action transfer,
remote 10.1.136.159@12443 QUIC cert-key
TFg9ybqubTukNtMiFdn5jW61Y4VUPS9XmYxHsCeQ/4c=
2026-01-06T10:28:03+0800 info: [foo.] IXFR, outgoing, remote
10.1.136.159@12443 QUIC, incomplete history, serial 2025123116, fallback
to AXFR
2026-01-06T10:28:03+0800 debug: [foo.] ACL, allowed, action transfer,
remote 10.1.136.159@12443 QUIC cert-key
TFg9ybqubTukNtMiFdn5jW61Y4VUPS9XmYxHsCeQ/4c=
2026-01-06T10:28:03+0800 info: [foo.] AXFR, outgoing, remote
10.1.136.159@12443 QUIC, started, serial 2025123118
2026-01-06T10:28:04+0800 info: [foo.] AXFR, outgoing, remote
10.1.136.159@12443 QUIC, buffering finished, 0.35 seconds, 3695
messages, 62246340 bytes
2026-01-06T10:28:08+0800 debug: QUIC, terminated inactive connections 1
from the master's log, the transfer to both slave is ok, followed by
"QUIC, terminated inactive connections 1",
but the slave2 don't get the zone.
-rw-r-----. 1 root root 1400373248 Jan 6 10:31
core.knotd.1000.eceeeff3d58f457db4014dc5f33e0fad.327676.1767666688000000
[root@tiangong-01 coredump]# gdb --core
core.knotd.1000.eceeeff3d58f457db4014dc5f33e0fad.327676.1767666688000000
GNU gdb (Rocky Linux) 16.3-2.el9
Copyright (C) 2024 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later
<http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
Type "show copying" and "show warranty" for details.
This GDB was configured as "aarch64-redhat-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<https://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word".
warning: Can't open file /tmp/knot-confdb.g2kMNg/lock.mdb (deleted)
during file-backed mapping note processing
warning: Can't open file /tmp/knot-confdb.g2kMNg/data.mdb (deleted)
during file-backed mapping note processing
[New LWP 327685]
[New LWP 327678]
[New LWP 327676]
[New LWP 327680]
[New LWP 327679]
[New LWP 327682]
[New LWP 327681]
[New LWP 327686]
[New LWP 327684]
[New LWP 327688]
[New LWP 327687]
[New LWP 327677]
[New LWP 327683]
[New LWP 327689]
[New LWP 327690]
Reading symbols from /usr/sbin/knotd...
Reading symbols from
/usr/lib/debug/usr/sbin/knotd-3.5.2-cznic.1.el9.aarch64.debug...
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib64/libthread_db.so.1".
Core was generated by `/usr/sbin/knotd -c
/home/gtld/knot/chroot4/etc/knot.conf -d'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0 __GI___libc_free (mem=0xa27d64bf12c00800) at malloc.c:3231
3231 if (chunk_is_mmapped (p)) /* release
mmapped memory. */
[Current thread is 1 (Thread 0xffff62c0dd00 (LWP 327685))]
Missing rpms, try: dnf --enablerepo='*debug*' install
libxdp-debuginfo-1.5.5-1.el9_6.aarch64
(gdb) bt
#0 __GI___libc_free (mem=0xa27d64bf12c00800) at malloc.c:3231
#1 0x0000ffff84a77c7c in _gnutls_buffer_clear
(str=str@entry=0xffff62c0bfc8) at str.c:80
#2 0x0000ffff84aaf4b8 in _gnutls13_recv_end_of_early_data
(session=session@entry=0xffff04032af0) at tls13/early_data.c:103
#3 0x0000ffff84a42d5c in _gnutls13_handshake_server
(session=session@entry=0xffff04032af0) at handshake-tls13.c:481
#4 0x0000ffff84a4ff8c in handshake_server (session=<optimized out>) at
handshake.c:3502
#5 gnutls_handshake (session=session@entry=0xffff04032af0) at
handshake.c:2898
#6 0x0000ffff850006a8 in ngtcp2_crypto_read_write_crypto_data
(conn=conn@entry=0xffff0402b780, encryption_level=<optimized out>,
data=<optimized out>,
datalen=<optimized out>) at
contrib/libngtcp2/ngtcp2/crypto/gnutls.c:498
#7 0x0000ffff8500070c in ngtcp2_crypto_recv_crypto_data_cb
(conn=0xffff0402b780, encryption_level=<optimized out>,
offset=<optimized out>,
data=<optimized out>, datalen=<optimized out>, user_data=<optimized
out>) at contrib/libngtcp2/ngtcp2/crypto/shared.c:1718
#8 0x0000ffff8500b104 in conn_call_recv_crypto_data (conn=<optimized
out>, encryption_level=<optimized out>, offset=<optimized out>,
data=<optimized out>, datalen=<optimized out>) at
contrib/libngtcp2/ngtcp2/lib/ngtcp2_conn.c:150
#9 0x0000ffff85023e2c in conn_emit_pending_crypto_data (rx_offset=935,
strm=0xffff04039dc0, encryption_level=NGTCP2_ENCRYPTION_LEVEL_INITIAL,
conn=0xffff0402b780) at contrib/libngtcp2/ngtcp2/lib/ngtcp2_conn.c:5968
#10 conn_recv_crypto.part.0.lto_priv.0 (conn=0xffff0402b780,
encryption_level=NGTCP2_ENCRYPTION_LEVEL_INITIAL, crypto=0xffff04039dc0,
fr=<optimized out>,
ts=<optimized out>) at contrib/libngtcp2/ngtcp2/lib/ngtcp2_conn.c:7281
#11 0x0000ffff8501a630 in conn_recv_crypto (ts=2855398215727159,
fr=0xffff62c0c310, crypto=0xffff04039dc0,
encryption_level=NGTCP2_ENCRYPTION_LEVEL_INITIAL,
conn=0xffff0402b780) at contrib/libngtcp2/ngtcp2/lib/ngtcp2_conn.c:7213
#12 conn_recv_handshake_pkt (conn=conn@entry=0xffff0402b780,
path=path@entry=0xffff62c0cd10, pi=pi@entry=0xffff62c0cd00,
pkt=pkt@entry=0xffff62164658 <incomplete sequence \303>,
pktlen=pktlen@entry=1200, dgramlen=dgramlen@entry=1200,
pkt_ts=pkt_ts@entry=2855398215727159, ts=ts@entry=2855398215727159)
at contrib/libngtcp2/ngtcp2/lib/ngtcp2_conn.c:6902
#13 0x0000ffff8501a984 in conn_recv_handshake_cpkt
(conn=conn@entry=0xffff0402b780, path=0xffff62c0cd10, pi=0xffff62c0cd00,
pkt=0xffff62164658 <incomplete sequence \303>,
pktlen=pktlen@entry=1200, ts=ts@entry=2855398215727159)
at contrib/libngtcp2/ngtcp2/lib/ngtcp2_conn.c:6987
#14 0x0000ffff8501b12c in conn_read_handshake (conn=0xffff0402b780,
path=<optimized out>, pi=<optimized out>, pkt=<optimized out>,
pktlen=1200,
ts=2855398215727159) at
contrib/libngtcp2/ngtcp2/lib/ngtcp2_conn.c:10091
#15 0x0000ffff85008498 in ngtcp2_conn_read_pkt_versioned
(pkt_info_version=1, ts=2855398215727159, pktlen=1200, pkt=<optimized
out>, pi=0xffff62c0cd00,
path=0xffff62c0cd10, conn=<optimized out>) at
contrib/libngtcp2/ngtcp2/lib/ngtcp2_conn.c:10308
#16 knot_quic_handle (table=table@entry=0xffff04000d40,
reply=reply@entry=0xffff62c0d060, idle_timeout=<optimized out>,
out_conn=out_conn@entry=0xffff62c0d008) at libknot/quic/quic.c:718
#17 0x0000aaaaba698244 in quic_handler (p_ecn=<optimized out>,
mh_out=0xffff62164298, rx=<optimized out>, table=0xffff04000d40,
idle_close=<optimized out>, layer=0xffff62c0d228,
params=0xffff62c0d010) at knot/server/quic-handler.c:90
#18 udp_mmsg_handle (ctx=0xffff62c0d228, iface=0xaaaad1487c10,
d=0xffff62164010) at knot/server/udp-handler.c:359
#19 0x0000aaaaba698ae4 in udp_master (thread=0xaaaad151b1d0) at
knot/server/udp-handler.c:624
#20 0x0000aaaaba68e330 in thread_ep (data=0xaaaad151b1d0) at
knot/server/dthreads.c:137
#21 0x0000ffff848d2e00 in start_thread (arg=0x10e540) at
pthread_create.c:443
#22 0x0000ffff8493d49c in thread_start () at
../sysdeps/unix/sysv/linux/aarch64/clone.S:79
(gdb)
gnutls 3.8.3-9.el9.0.1 baseos
ngtcp2 1.15.1-2.el9 epel
Best Regards,
SUN Guonian
Hi,
On 1/5/26 10:54, SUN Guonian wrote:
2026-01-06T10:24:38+0800 debug: [foo.] ACL, allowed, action
notify,
remote 10.1.136.154@21789 TCP
2026-01-06T10:24:38+0800 info: [foo.] notify, incoming, remote
10.1.136.154@21789 TCP, serial 2025123118
2026-01-06T10:24:38+0800 info: [foo.] refresh, remote 10.1.136.154@853,
remote serial 2025123118, zone is outdated
2026-01-06T10:24:39+0800 info: [foo.] IXFR, incoming, remote
10.1.136.154@853 QUIC/0-RTT, receiving AXFR-style IXFR
2026-01-06T10:24:39+0800 info: [foo.] AXFR, incoming, remote
10.1.136.154@853 QUIC/0-RTT, started
2026-01-06T10:24:42+0800 info: [foo.] AXFR, incoming, remote
10.1.136.154@853 QUIC, finished, remote serial 2025123118, 2.99 seconds,
3695 messages, 62246340 bytes
2026-01-06T10:24:43+0800 info: [foo.] refresh, remote 10.1.136.154@853,
zone updated, 4.82 seconds, serial 2025123117 -> 2025123118, expires in
2491200 seconds
2026-01-06T10:24:45+0800 info: [foo.] zone file updated, serial
2025123117 -> 2025123118
2026-01-06T10:33:13+0800 debug: stats, dumped into file
'/home/gtld/knot/chroot1/var/run/stats.yaml'
2026-01-06T10:34:44+0800 info: [foo.] refresh, remote master_quic,
address 10.1.136.154@853, failed (connection reset)
2026-01-06T10:34:44+0800 warning: [foo.] refresh, remote master_quic not
usable
2026-01-06T10:34:44+0800 error: [foo.] refresh, failed (no usable
master), next retry at 2026-01-06T10:38:04+0800, expires in 2490599 seconds
2026-01-06T10:34:44+0800 error: [foo.] zone event 'refresh' failed (no
usable master)
在 2026/1/5 16:30, Daniel Salzman via knot-dns-users 写道:
This is interesting, because the number of background workers
shouldn't matter. The output buffer limit is per worker.
>
>>
>>>
>>> 1. I increase quic-outbuf-max-size from default 100M to 3200M,
>>> doubled each time, it still produce,
>>>
>>> 2026-01-05T14:47:49+0800 info: [foo.] AXFR, outgoing, remote
>>> 10.1.136.156@37936 QUIC, started, serial 2025123113
>>> 2026-01-05T14:47:49+0800 info: [foo.] AXFR, outgoing, remote
>>> 10.1.136.156@37936 QUIC, buffering finished, 0.27 seconds, 2956
>>> messages, 49797072 bytes
>>> 2026-01-05T14:49:13+0800 debug: QUIC, terminated inactive
>>> connections 1
>>> 2026-01-05T14:49:13+0800 debug: QUIC, terminated inactive
>>> connections 1
>>> 2026-01-05T14:49:19+0800 debug: [foo.] ACL, allowed, action
>>> transfer, remote 10.1.136.159@14324 QUIC cert-key
>>> TFg9ybqubTukNtMiFdn5jW61Y4VUPS9XmYxHsCeQ/4c=
>>> 2026-01-05T14:49:19+0800 info: [foo.] AXFR, outgoing, remote
>>> 10.1.136.159@14324 QUIC, started, serial 2025123113
>>> 2026-01-05T14:49:19+0800 info: [foo.] AXFR, outgoing, remote
>>> 10.1.136.159@14324 QUIC, buffering finished, 0.27 seconds, 2956
>>> messages, 49797072 bytes
>>> 2026-01-05T14:49:23+0800 debug: QUIC, terminated inactive
>>> connections 1
>>
>> Note that the reason for connection termination is different. The
>> client was inactive.
>>
>>>
>>> 2. one slave(10.1.136.156, rocky9) could get the zone fully,
>>> another (10.1.136.159, rock10) couldn't, both are alive.
>>
>> What does the server log?
>
> on slave1,
>
> 2026-01-05T17:31:58+0800 info: [foo.] notify, incoming, remote
> 10.1.136.154@24037 TCP, serial 2025123116
> 2026-01-05T17:31:58+0800 info: [foo.] refresh, remote
> 10.1.136.154@853, remote serial 2025123116, zone is outdated
> 2026-01-05T17:31:58+0800 info: [foo.] IXFR, incoming, remote
> 10.1.136.154@853 QUIC, receiving AXFR-style IXFR
> 2026-01-05T17:31:58+0800 info: [foo.] AXFR, incoming, remote
> 10.1.136.154@853 QUIC, started
> 2026-01-05T17:32:02+0800 info: [foo.] AXFR, incoming, remote
> 10.1.136.154@853 QUIC, finished, remote serial 2025123116, 3.42
> seconds, 3695 messages, 62246340 bytes
> 2026-01-05T17:32:03+0800 info: [foo.] refresh, remote
> 10.1.136.154@853, zone updated, 5.56 seconds, serial 2025123114 ->
> 2025123116, expires in 2491200 seconds
> 2026-01-05T17:32:05+0800 info: [foo.] zone file updated, serial
> 2025123114 -> 2025123116
> 2026-01-05T17:33:13+0800 debug: stats, dumped into file
> '/home/gtld/knot/chroot1/var/run/stats.yaml'
> 2026-01-05T17:42:08+0800 info: [foo.] refresh, remote master_quic,
> address 10.1.136.154@853, failed (connection reset)
> 2026-01-05T17:42:08+0800 warning: [foo.] refresh, remote master_quic
> not usable
> 2026-01-05T17:42:08+0800 error: [foo.] refresh, failed (no usable
> master), next retry at 2026-01-05T17:45:28+0800, expires in 2490595
> seconds
> 2026-01-05T17:42:08+0800 error: [foo.] zone event 'refresh' failed
> (no usable master)
> 2026-01-05T17:45:30+0800 info: [foo.] refresh, remote master_quic,
> address 10.1.136.154@853, failed (connection reset)
> 2026-01-05T17:45:30+0800 warning: [foo.] refresh, remote master_quic
> not usable
> 2026-01-05T17:45:30+0800 error: [foo.] refresh, failed (no usable
> master), next retry at 2026-01-05T17:48:50+0800, expires in 2490393
> seconds
> 2026-01-05T17:45:30+0800 error: [foo.] zone event 'refresh' failed
> (no usable master) Hi!
On 1/5/26 08:06, SUN Guonian wrote:
background worker is the default value,
on x86_64, it is 2 for both master & slave.
on aarch64, master is 48, slave1(10.1.136.156) is 48,
slave2(10.1.136.159, virtual host) is 24.
I have decreased it to 2 on the master, both slave could get the zone. It take effect to increase quic-outbuf-max-size,
the transfer works
on architecture x86_64.
but on aarch64, there is a lot difference,
Are there the same number of background workers on both architectures? >
> on slave2,
>
> 2026-01-05T17:32:28+0800 info: server started
> 2026-01-05T17:32:28+0800 info: [foo.] AXFR, incoming, remote
> 10.1.136.154@853 QUIC, started
> 2026-01-05T17:32:33+0800 info: [foo.] AXFR, incoming, remote
> 10.1.136.154@853 QUIC, finished, remote serial 2025123116, 4.94
> seconds, 3695 messages, 62246340 bytes
> 2026-01-05T17:32:35+0800 info: [foo.] refresh, remote
> 10.1.136.154@853, zone updated, 7.34 seconds, serial none ->
> 2025123116, expires in 2491200 seconds
> 2026-01-05T17:32:37+0800 info: [foo.] zone file updated, serial
> 2025123116
> 2026-01-05T17:34:59+0800 debug: [foo.] ACL, allowed, action notify,
> remote 10.1.136.154@21413 TCP
> 2026-01-05T17:34:59+0800 info: [foo.] notify, incoming, remote
> 10.1.136.154@21413 TCP, serial 2025123116
> 2026-01-05T17:42:36+0800 info: [foo.] refresh, remote master_quic,
> address 10.1.136.154@853, failed (connection reset)
> 2026-01-05T17:42:36+0800 warning: [foo.] refresh, remote master_quic
> not usable
> 2026-01-05T17:42:36+0800 error: [foo.] refresh, failed (no usable
> master), next retry at 2026-01-05T17:45:56+0800, expires in 2490599
> seconds
> 2026-01-05T17:42:36+0800 error: [foo.] zone event 'refresh' failed
> (no usable master)
> 2026-01-05T17:45:58+0800 info: [foo.] refresh, remote master_quic,
> address 10.1.136.154@853, failed (connection reset)
> 2026-01-05T17:45:58+0800 warning: [foo.] refresh, remote master_quic
> not usable
> 2026-01-05T17:45:58+0800 error: [foo.] refresh, failed (no usable
> master), next retry at 2026-01-05T17:49:18+0800, expires in 2490397
> seconds
> 2026-01-05T17:45:58+0800 error: [foo.] zone event 'refresh' failed
> (no usable master)
2026-01-06T10:27:44+0800 info: [foo.] refresh, remote
10.1.136.154@853,
remote serial 2025123118, zone is outdated
2026-01-06T10:27:49+0800 info: [foo.] refresh, remote master_quic,
address 10.1.136.154@853, failed (connection reset)
2026-01-06T10:27:49+0800 warning: [foo.] refresh, remote master_quic not
usable
2026-01-06T10:27:49+0800 error: [foo.] refresh, failed (no usable
master), next retry at 2026-01-06T10:31:09+0800, expires in 2431693 seconds
2026-01-06T10:27:49+0800 error: [foo.] zone event 'refresh' failed (no
usable master)
2026-01-06T10:31:12+0800 info: [foo.] refresh, remote master_quic,
address 10.1.136.154@853, failed (connection reset)
2026-01-06T10:31:12+0800 warning: [foo.] refresh, remote master_quic not
usable
2026-01-06T10:31:12+0800 error: [foo.] refresh, failed (no usable
master), next retry at 2026-01-06T10:34:32+0800, expires in 2431490 seconds
2026-01-06T10:31:12+0800 error: [foo.] zone event 'refresh' failed (no
usable master)
After slave2 do the second try, the process on master disappear.
on master,
2026-01-05T17:31:58+0800 info: server started
2026-01-05T17:31:58+0800 debug: [foo.] ACL, allowed, action transfer,
remote 10.1.136.156@45612 QUIC cert-key
kulz9ehQf5Ycn/+2mCicUdfTMuDXHbQEWBwg5qDi0Eo=
2026-01-05T17:31:58+0800 info: [foo.] IXFR, outgoing, remote
10.1.136.156@45612 QUIC, incomplete history, serial 2025123114,
fallback to AXFR
2026-01-05T17:31:58+0800 debug: [foo.] ACL, allowed, action transfer,
remote 10.1.136.156@45612 QUIC cert-key
kulz9ehQf5Ycn/+2mCicUdfTMuDXHbQEWBwg5qDi0Eo=
2026-01-05T17:31:58+0800 info: [foo.] AXFR, outgoing, remote
10.1.136.156@45612 QUIC, started, serial 2025123116
2026-01-05T17:31:58+0800 info: [foo.] AXFR, outgoing, remote
10.1.136.156@45612 QUIC, buffering finished, 0.33 seconds, 3695
messages, 62246340 bytes
2026-01-05T17:32:47+0800 debug: [foo.] ACL, allowed, action transfer,
remote 10.1.136.159@41166 QUIC cert-key
TFg9ybqubTukNtMiFdn5jW61Y4VUPS9XmYxHsCeQ/4c=
2026-01-05T17:32:47+0800 info: [foo.] AXFR, outgoing, remote
10.1.136.159@41166 QUIC, started, serial 2025123116
2026-01-05T17:32:47+0800 info: [foo.] AXFR, outgoing, remote
10.1.136.159@41166 QUIC, buffering finished, 0.38 seconds, 3695
messages, 62246340 bytes
2026-01-05T17:35:18+0800 info: [foo.] notify, outgoing, remote
10.1.136.159@53 TCP, retry, serial 2025123116
These logs describe different situation. I see successful transfers
over QUIC. Note that the failed transfers were over TCP.
master still crashed,
-rw-r-----. 1 root root 65663813 Jan 5 17:42
core.knotd.1000.eceeeff3d58f457db4014dc5f33e0fad.323451.1767606123000000.zst
this command output,
coredumpctl: unrecognized option '--batch'
I try to get the output for gdb/bt, but failed,
# coredumpctl dump > c01
PID: 323451 (knotd)
UID: 1000 (gtld)
GID: 1000 (gtld)
Signal: 11 (SEGV)
Timestamp: Mon 2026-01-05 17:42:03 CST (6min ago)
Command Line: /usr/sbin/knotd -c
/home/gtld/knot/chroot4/etc/knot.conf -d
Executable: /usr/sbin/knotd
Control Group: /user.slice/user-0.slice/session-129.scope
Unit: session-129.scope
Slice: user-0.slice
Session: 129
Owner UID: 0 (root)
Boot ID: eceeeff3d58f457db4014dc5f33e0fad
Machine ID: 82aa697a9ae54202bb5e0ec31c510520
Hostname: tiangong-01
Storage:
/var/lib/systemd/coredump/core.knotd.1000.eceeeff3d58f457db4014dc5f33e0fad.323451.1767606123000000.zst
(truncated)
Size on Disk: 62.6M
Message: Process 323451 (knotd) of user 1000 dumped core.
Stack trace of thread 323458:
#0 0x0000ffff9bce36f0 n/a (n/a + 0x0)
#1 0x0000ffff9be77c7c n/a (n/a + 0x0)
#2 0x0000ffff9be77c7c n/a (n/a + 0x0)
#3 0x0000ffff9beaf4b8 n/a (n/a + 0x0)
#4 0x0000ffff9be42d5c n/a (n/a + 0x0)
#5 0x0000ffff9be4ff8c n/a (n/a + 0x0)
#6 0x0000ffff9c3516a8 n/a (n/a + 0x0)
#7 0x0000ffff9c35170c n/a (n/a + 0x0)
#8 0x0000ffff9c35c104 n/a (n/a + 0x0)
#9 0x0000ffff9c374e2c n/a (n/a + 0x0)
#10 0x0000ffff9c36b630 n/a (n/a + 0x0)
#11 0x0000ffff9c36b984 n/a (n/a + 0x0)
#12 0x0000ffff9c36c12c n/a (n/a + 0x0)
#13 0x0000ffff9c359498 n/a (n/a + 0x0)
#14 0x0000aaaae3cf8244 quic_handler (/usr/sbin/knotd
+ 0x28244)
#15 0xeee58b05aa78eb00 n/a (n/a + 0x0)
ELF object binary architecture: AARCH64
More than one entry matches, ignoring rest.
# gdb --core c01
GNU gdb (Rocky Linux) 16.3-2.el9
Copyright (C) 2024 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later
<http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.
Type "show copying" and "show warranty" for details.
This GDB was configured as "aarch64-redhat-linux-gnu".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<https://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word".
warning: BFD: warning: /var/lib/systemd/coredump/c01 has a segment
extending past end of file
warning: Can't open file /tmp/knot-confdb.B4NsUg/lock.mdb (deleted)
during file-backed mapping note processing
warning: Can't open file /tmp/knot-confdb.B4NsUg/data.mdb (deleted)
during file-backed mapping note processing
[New LWP 323458]
[New LWP 323453]
[New LWP 323451]
[New LWP 323454]
[New LWP 323456]
[New LWP 323460]
[New LWP 323455]
[New LWP 323461]
[New LWP 323457]
[New LWP 323459]
[New LWP 323462]
[New LWP 323463]
[New LWP 323452]
[New LWP 323464]
[New LWP 323465]
warning: failed to parse execution context from corefile: Cannot
access memory at address 0xffffc31bcfe8
Reading symbols from /usr/sbin/knotd...
Reading symbols from
/usr/lib/debug/usr/sbin/knotd-3.5.2-cznic.1.el9.aarch64.debug...
warning: Error reading shared library list entry at 0x2e20613635303264
Cannot access memory at address 0x6578652d6c642f80
Cannot access memory at address 0x6578652d6c642f78
Failed to read a valid object file image from memory.
Core was generated by `/usr/sbin/knotd -c
/home/gtld/knot/chroot4/etc/knot.conf -d'.
Program terminated with signal SIGSEGV, Segmentation fault.
#0 0x0000ffff9bce36f0 in ?? ()
[Current thread is 1 (LWP 323458)]
(gdb) bt
#0 0x0000ffff9bce36f0 in ?? ()
Backtrace stopped: previous frame identical to this frame (corrupt
stack?)
(gdb)
Unfortunately, this output is incomplete. You should increase the
curedump limit and install packages with debug symbols:
knot-debuginfo knot-libs-debuginfo
3. the master(10.1.136.154, rocky9) collapsed.
# ls -lrt /var/lib/systemd/coredump/
-rw-r-----. 1 root root 52527031 Jan 5 09:09
core.knotd.1000.eceeeff3d58f457db4014dc5f33e0fad.320376.1767575365000000.zst
-rw-r-----. 1 root root 52575905 Jan 5 10:06
core.knotd.1000.eceeeff3d58f457db4014dc5f33e0fad.320623.1767578803000000.zst
-rw-r-----. 1 root root 52546840 Jan 5 10:53
core.knotd.1000.eceeeff3d58f457db4014dc5f33e0fad.320970.1767581613000000.zst
-rw-r-----. 1 root root 52519031 Jan 5 13:33
core.knotd.1000.eceeeff3d58f457db4014dc5f33e0fad.321756.1767591216000000.zst
-rw-r-----. 1 root root 52519386 Jan 5 13:46
core.knotd.1000.eceeeff3d58f457db4014dc5f33e0fad.321896.1767591963000000.zst
-rw-r-----. 1 root root 52512650 Jan 5 14:23
core.knotd.1000.eceeeff3d58f457db4014dc5f33e0fad.322020.1767594182000000.zst
-rw-r-----. 1 root root 52553042 Jan 5 14:49
core.knotd.1000.eceeeff3d58f457db4014dc5f33e0fad.322309.1767595778000000.zst
Could you send me a backtrace? Something like
coredumpctl gdb knot \
--batch \
-ex "set pagination off" \
-ex "thread apply all bt full"
Daniel
>
> Thanks!
>
>
> Best Regards,
>
> SUN Guonian
>
>>
>> Thanks!
>>
>>>
>>> Thanks !
>>>
>>> Best Regards,
>>>
>>> SUN Guonian
>>>
>>> 在 2026/1/4 19:05, Daniel Salzman 写道:
>>>> Hello!
>>>>
>>>> You probably have to increase
>>>>
https://www.knot-dns.cz/docs/latest/singlehtml/index.html#quic-outbuf-max-s…
>>>>
>>>> Daniel
>>>>
>>>> On 1/4/26 10:46, SUN Guonian via knot-dns-users wrote:
>>>>> Greetings,
>>>>>
>>>>> I have tried to use QUIC in zone transfering, I met one error in
>>>>> on bigger zone,
>>>>>
>>>>> from master's log, it displayed,
>>>>> 2026-01-04T17:32:19+0800 debug: [foo.] ACL, allowed, action
>>>>> transfer, remote 10.0.0.147@60880 QUIC cert-key
>>>>> xJKsDkUqpl6orXeTwsrDgDvgZ/PiYxOSVlOkVdn5EOU=
>>>>> 2026-01-04T17:32:19+0800 info: [foo.] IXFR, outgoing, remote
>>>>> 10.0.0.147@60880 QUIC, incomplete history, serial 2026010403,
>>>>> fallback to AXFR
>>>>> 2026-01-04T17:32:19+0800 debug: [foo.] ACL, allowed, action
>>>>> transfer, remote 10.0.0.147@60880 QUIC cert-key
>>>>> xJKsDkUqpl6orXeTwsrDgDvgZ/PiYxOSVlOkVdn5EOU=
>>>>> 2026-01-04T17:32:19+0800 info: [foo.] AXFR, outgoing, remote
>>>>> 10.0.0.147@60880 QUIC, started, serial 2026010404
>>>>> 2026-01-04T17:32:20+0800 info: [foo.] AXFR, outgoing, remote
>>>>> 10.0.0.147@60880 QUIC, buffering finished, 0.87 seconds, 7390
>>>>> messages, 124493148 bytes
>>>>> 2026-01-04T17:32:20+0800 notice: QUIC, terminated connections,
>>>>> outbuf limit 1
>>>>>
>>>>> on the slave side, I got log as,
>>>>> 2026-01-04T17:32:18+0800 info: [foo.] zone file loaded, serial
>>>>> 2026010403
>>>>> 2026-01-04T17:32:19+0800 info: [foo.] loaded, serial none ->
>>>>> 2026010403, 92000117 bytes
>>>>> 2026-01-04T17:32:19+0800 info: [foo.] refresh, remote
>>>>> 10.0.0.151@853, remote serial 2026010404, zone is outdated
>>>>> 2026-01-04T17:32:19+0800 info: server started
>>>>>
>>>>> (and, the knotd on slave will down without log.)
>>>>>
>>>>> Thanks in advance.
>>>>>
>>>>>
>>>>> My testing environment is,
>>>>>
>>>>> the zone size is 1,000,000 x ( 2 NS + 2 A ), such as,
>>>>> domain00000000 3600 NS ns1.domain00000000
>>>>> 3600 NS ns2.domain00000000
>>>>> ns1.domain00000000 3600 A 10.0.0.1
>>>>> ns2.domain00000000 3600 A 10.0.0.2
>>>>> ...
>>>>> domain00999999 3600 NS ns1.domain00999999
>>>>> 3600 NS ns2.domain00999999
>>>>> ns1.domain00999999 3600 A 10.0.0.1
>>>>> ns2.domain00999999 3600 A 10.0.0.2
>>>>>
>>>>> If I decrease the record number to 500,000 x ( 2 NS + 2 A ), the
>>>>> zone could be transfer with QUIC successfully.
>>>>>
>>>>> For traditional TCP and TLS, the zone transfer is processed
>>>>> without error, even for more large size.
>>>>>
>>>>> Version in master and slave are both 3.5.2, installed from copr.
>>>>> OS in both side is Rocky9 x86_64.
>>>>>
>>>>> Best Regards,
>>>>> SUN Guonian
>>>>>
>>>>> --