Package: ser2net
Version: 3.5-2
Severity: normal
Dear Maintainer,
I have been using ser2net to expose a device connected a serial port of
microPC.
As the original host application was showing some spurious errors on
large transfers,
I enabled tracelogs (see config below).
The tracelogs showed strange behaviour: On a certain command, the device
should have responded
with a sequence of 2 replies, but the tracelogs showed basically a
repeat of the first reply.
I assume this is due to the logging function looking at the beginning of
the to-be-sent buffer,
as the reply happened to include a \377 (which is the telnet escape
char) which was duplicated.
Excerpt from strace:
# command below should elicit a 2x5 byte response, starting with 0x35
0x1c and 0x35 0x1b
read(0, "5\32\350\0S", 64) = 5
write(2, "2021/09/14 14:27:18 tcp 35 1a e8 00 53 |5...S|\n",
58) = 58
write(1, "5\32\350\0S", 5) = 5
# so far correct
read(1, "5\34\377\377S", 32) = 5
write(2, "2021/09/14 14:27:18 term 35 1c ff ff 53 |5...S|\n",
58) = 58
# device answers, first packet, logging is correct
--- SIGUSR1 {si_signo=SIGUSR1, si_code=SI_TKILL, si_pid=21545, si_uid=0} ---
read(1, "5\33\0\4S", 28) = 5
write(2, "2021/09/14 14:27:18 term 35 1c ff ff ff |5....|\n",
58) = 58
# device answers, second packet. Logging is wrong. Should be 35 1b 00 04 53
# note that we actually see the first packet, and the last 53 is
replaced by ff
# I assume this is due to the already-escaped buffer below being logged
from the
# beginning:
--- SIGUSR1 {si_signo=SIGUSR1, si_code=SI_TKILL, si_pid=21545, si_uid=0} ---
--- SIGUSR1 {si_signo=SIGUSR1, si_code=SI_TKILL, si_pid=21545, si_uid=0} ---
write(0, "5\34\377\377\377\377S5\33\0\4S", 12) = 12
# both answers go over the network in one packet, doubling the \377 = ff
is correct due to telnet escape rules
Unfortunately cannot update the microPC to stable to verify this is not
in the most recent version.
Functionality of the packet is not affected, but it cost me several
hours of debugging time to find the
logging is incorrect, as opposed to the device malfunctioning.
Thanks for your efforts,
Andreas Beck
-- System Information:
Debian Release: 10.10
APT prefers oldstable-updates
APT policy: (500, 'oldstable-updates'), (500, 'oldstable')
Architecture: amd64 (x86_64)
Kernel: Linux 4.19.0-17-amd64 (SMP w/4 CPU cores)
Kernel taint flags: TAINT_WARN, TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8),
LANGUAGE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled
Versions of packages ser2net depends on:
ii libc6 2.28-10
ii libwrap0 7.6.q-28
ii lsb-base 10.2019051400
ser2net recommends no packages.
Versions of packages ser2net suggests:
ii telnet 0.17-41.2
-- Configuration Files:
/etc/ser2net.conf changed:
BANNER:banner:\r\nser2net port \p device \d [\s] (Debian GNU/Linux)\r\n\r\n
TRACEFILE:trace:/tmp/ser2net\p.log
4711:telnet:600:/dev/ttyUSB0:38400 8DATABITS NONE 1STOPBIT tb=trace
hexdump timestamp
4712:telnet:600:/dev/pts/0:38400 8DATABITS NONE 1STOPBIT tb=trace
hexdump timestamp
-- no debconf information
