Package: bind9-host
Version: 1:9.9.5.dfsg-12.1
Followup-For: Bug #548096
Dear Maintainer,
An earlier respondent mentioned the hang occuring with the "host" command as
well.
I have an Alpha PWS-433au running 4.5.0-rc4 (compiled from the stock
"kernel.org"
source tree). Libraries and userspace are fully up-to-date "sid". I've noticed
the hang issue since approximately the beginning of February of this year.
Locally, running "strace host gherkin.frus.com" (gherkin is my local nameserver)
runs to completion about half the time. The rest of the time, I see this:
(...)
futex(0x20002e6fad4, FUTEX_WAKE_OP_PRIVATE, 1, 1, 0x20002e6fad0, {FUTEX_OP_SET,
0, FUTEX_OP_CMP_GT, 1}) = 1
futex(0x20002e6fa58, FUTEX_WAKE_PRIVATE, 1) = 1
rt_sigaction(SIGHUP, {0x2000077b100, ~[RTMIN RT_1], 0}, NULL, 8, 0x200007ff0c0)
= 0
sigsuspend([FPE]gherkin.frus.com has address 192.158.254.49
gherkin.frus.com has IPv6 address 2001:5c0:1500:a800:222:4dff:fe51:6246
gherkin.frus.com mail is handled by 10 merkin.frus.com.
gherkin.frus.com mail is handled by 15 gherkin.frus.com.
gherkin.frus.com mail is handled by 21 mx-b.cirr.com.
gherkin.frus.com mail is handled by 20 mx-a.cirr.com.
A successful run looks like this:
(...)
futex(0x20002e6fad4, FUTEX_WAKE_OP_PRIVATE, 1, 1, 0x20002e6fad0, {FUTEX_OP_SET,
0, FUTEX_OP_CMP_GT, 1}) = 1
futex(0x20002e6fa58, FUTEX_WAKE_PRIVATE, 1) = 1
rt_sigaction(SIGHUP, {0x2000077b100, ~[RTMIN RT_1], 0}, NULL, 8, 0x200007ff0c0)
= 0
sigsuspend([FPE]gherkin.frus.com has address 192.158.254.49
gherkin.frus.com has IPv6 address 2001:5c0:1500:a800:222:4dff:fe51:6246
gherkin.frus.com mail is handled by 20 mx-a.cirr.com.
gherkin.frus.com mail is handled by 10 merkin.frus.com.
gherkin.frus.com mail is handled by 21 mx-b.cirr.com.
gherkin.frus.com mail is handled by 15 gherkin.frus.com.
) = ? ERESTARTNOHAND (To be restarted if no handler)
--- SIGTERM {si_signo=SIGTERM, si_code=SI_TKILL, si_pid=4925, si_uid=1000} ---
sigreturn({mask=[HUP INT TERM]}) = -1 EINTR (Interrupted system call)
futex(0x2000102f9c0, FUTEX_WAIT_PRIVATE, 2, NULL) = 0
futex(0x2000102f9c0, FUTEX_WAKE_PRIVATE, 1) = 0
futex(0x20002e6fad4, FUTEX_WAKE_OP_PRIVATE, 1, 1, 0x20002e6fad0, {FUTEX_OP_SET,
0, FUTEX_OP_CMP_GT, 1}) = 1
futex(0x20002e6fa58, FUTEX_WAKE_PRIVATE, 1) = 1
futex(0x20002e6fad4, FUTEX_CMP_REQUEUE_PRIVATE, 1, 2147483647, 0x20002e6fa58,
12) = 1
futex(0x20002e6fa58, FUTEX_WAKE_PRIVATE, 1) = 1
write(4, "\0\0\0\0\377\377\377\377", 8) = 8
epoll_ctl(5, EPOLL_CTL_DEL, 3, 11fb63430) = 0
close(5) = 0
close(3) = 0
close(4) = 0
futex(0x20002e71aac, FUTEX_WAKE_OP_PRIVATE, 1, 1, 0x20002e71aa8, {FUTEX_OP_SET,
0, FUTEX_OP_CMP_GT, 1}) = 1
futex(0x20002e71a58, FUTEX_WAKE_PRIVATE, 1) = 1
open("/proc/sys/vm/overcommit_memory", O_RDONLY|O_CLOEXEC) = 3
read(3, "0", 1) = 1
close(3) = 0
madvise(0x20008022000, 196608, MADV_DONTNEED) = 0
munmap(0x20000038000, 177304) = 0
exit_group(0) = ?
+++ exited with 0 +++
This "smells" like a race condition of some kind, but I have no idea where
the issue really lies. Could be something changed in one of the recent 4.X
kernels, or could be a library update that triggered this. The "bind9" packages
have been fairly stable, and I don't recall an update to them prior to this
issue appearing.
Respectfully,
--Bob
-- System Information:
Debian Release: stretch/sid
APT prefers unreleased
APT policy: (500, 'unreleased'), (500, 'unstable')
Architecture: alpha
Kernel: Linux 4.5.0-rc4
Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash
Init: systemd (via /run/systemd/system)
Versions of packages bind9-host depends on:
ii libbind9-90 1:9.9.5.dfsg-12.1
ii libc6.1 2.22-3
ii libcap2 1:2.24-12
ii libcomerr2 1.42.13-1
ii libdns100 1:9.9.5.dfsg-12.1
ii libgssapi-krb5-2 1.13.2+dfsg-5
ii libisc95 1:9.9.5.dfsg-12.1
ii libisccfg90 1:9.9.5.dfsg-12.1
ii libk5crypto3 1.13.2+dfsg-5
ii libkrb5-3 1.13.2+dfsg-5
ii liblwres90 1:9.9.5.dfsg-12.1
ii libssl1.0.2 1.0.2g-1
ii libxml2 2.9.3+dfsg1-1
bind9-host recommends no packages.
bind9-host suggests no packages.
-- no debconf information
