managesieve-crash on 2.4.2
Hi,
while trying to connect with "sieve-connect", managesieve crashed:
Nov 25 14:21:08 mail.schafweide.org dovecot[440904]: managesieve-login: Panic: file client.c: line 350 (managesieve_client_input_next_cmd): assertion failed: (args != NULL) Nov 25 14:21:08 mail.schafweide.org systemd-coredump[440938]: [🡕] Process 440934 (managesieve-log) of user 74 dumped core.
Stack trace of thread 440934: #0 0x000079c6cfa9890c n/a (libc.so.6 + 0x9890c) #1 0x000079c6cfa3e3a0 raise (libc.so.6 + 0x3e3a0) #2 0x000079c6cfa2557a abort (libc.so.6 + 0x2557a) #3 0x000079c6cfe5ffff n/a (libdovecot.so.0 + 0x5ffff) #4 0x000079c6cff4f095 n/a (libdovecot.so.0 + 0x14f095) #5 0x000079c6cfe5feb6 i_panic (libdovecot.so.0 + 0x5feb6) #6 0x000062e017df80de n/a (/usr/lib/dovecot/managesieve-login + 0x30de) #7 0x000062e017df8887 n/a (/usr/lib/dovecot/managesieve-login + 0x3887) #8 0x000079c6cff681f4 io_loop_call_io (libdovecot.so.0 + 0x1681f4) #9 0x000079c6cff6a2db io_loop_handler_run_internal (libdovecot.so.0 + 0x16a2db) #10 0x000079c6cff6a397 io_loop_handler_run (libdovecot.so.0 + 0x16a397) #11 0x000079c6cff6a599 io_loop_run (libdovecot.so.0 + 0x16a599) #12 0x000079c6cfeabce1 master_service_run (libdovecot.so.0 + 0xabce1) #13 0x000079c6d02f4350 login_binary_run (libdovecot-login.so.0 + 0x14350) #14 0x000079c6cfa27635 n/a (libc.so.6 + 0x27635) #15 0x000079c6cfa276e9 __libc_start_main (libc.so.6 + 0x276e9) #16 0x000062e017df8405 n/a (/usr/lib/dovecot/managesieve-login + 0x3405) ELF object binary architecture: AMD x86-64 Nov 25 14:21:08 mail.schafweide.org dovecot[440904]: managesieve-login: Fatal: master: service(managesieve-login): child 440934 killed with signal 6 (core dumped)
I'm using dovecot/pigeonhole 2.4.2 on Arch.
Regards Bjoern
Hi,
It would be helpful if you could obtain a rawlog of the connection:
https://doc.dovecot.org/2.4.2/core/admin/rawlog.html#pre-login-rawlog
Also, you need to install (or compile with) debug symbols to obtain a more useful backtrace.
Regards,
Stephan.
Op 25-11-2025 om 14:51 schreef Bjoern Franke via dovecot:
Hi,
while trying to connect with "sieve-connect", managesieve crashed:
Nov 25 14:21:08 mail.schafweide.org dovecot[440904]: managesieve-login: Panic: file client.c: line 350 (managesieve_client_input_next_cmd): assertion failed: (args != NULL) Nov 25 14:21:08 mail.schafweide.org systemd-coredump[440938]: [🡕] Process 440934 (managesieve-log) of user 74 dumped core.
Stack trace of thread 440934: #0 0x000079c6cfa9890c n/a (libc.so.6 + 0x9890c) #1 0x000079c6cfa3e3a0 raise (libc.so.6 + 0x3e3a0) #2 0x000079c6cfa2557a abort (libc.so.6 + 0x2557a) #3 0x000079c6cfe5ffff n/a (libdovecot.so.0 + 0x5ffff) #4 0x000079c6cff4f095 n/a (libdovecot.so.0 + 0x14f095) #5 0x000079c6cfe5feb6 i_panic (libdovecot.so.0 + 0x5feb6) #6 0x000062e017df80de n/a (/usr/lib/dovecot/managesieve-login + 0x30de) #7 0x000062e017df8887 n/a (/usr/lib/dovecot/managesieve-login + 0x3887) #8 0x000079c6cff681f4 io_loop_call_io (libdovecot.so.0 + 0x1681f4) #9 0x000079c6cff6a2db io_loop_handler_run_internal (libdovecot.so.0 + 0x16a2db) #10 0x000079c6cff6a397 io_loop_handler_run (libdovecot.so.0 + 0x16a397) #11 0x000079c6cff6a599 io_loop_run (libdovecot.so.0 + 0x16a599) #12 0x000079c6cfeabce1 master_service_run (libdovecot.so.0 + 0xabce1) #13 0x000079c6d02f4350 login_binary_run (libdovecot-login.so.0 + 0x14350) #14 0x000079c6cfa27635 n/a (libc.so.6 + 0x27635) #15 0x000079c6cfa276e9 __libc_start_main (libc.so.6 + 0x276e9) #16 0x000062e017df8405 n/a (/usr/lib/dovecot/managesieve-login + 0x3405) ELF object binary architecture: AMD x86-64 Nov 25 14:21:08 mail.schafweide.org dovecot[440904]: managesieve-login: Fatal: master: service(managesieve-login): child 440934 killed with signal 6 (core dumped)
I'm using dovecot/pigeonhole 2.4.2 on Arch.
Regards Bjoern
dovecot mailing list -- dovecot@dovecot.org To unsubscribe send an email to dovecot-leave@dovecot.org
Op 25-11-2025 om 15:26 schreef Stephan Bosch via dovecot:
Hi,
It would be helpful if you could obtain a rawlog of the connection:
https://doc.dovecot.org/2.4.2/core/admin/rawlog.html#pre-login-rawlog
Also, you need to install (or compile with) debug symbols to obtain a more useful backtrace.
Oh, and you need to (temporarily!) run managesieve-login without chroot as well to get a proper backtrace.
Regards,
Stephan.
Op 25-11-2025 om 14:51 schreef Bjoern Franke via dovecot:
Hi,
while trying to connect with "sieve-connect", managesieve crashed:
Nov 25 14:21:08 mail.schafweide.org dovecot[440904]: managesieve-login: Panic: file client.c: line 350 (managesieve_client_input_next_cmd): assertion failed: (args != NULL) Nov 25 14:21:08 mail.schafweide.org systemd-coredump[440938]: [🡕] Process 440934 (managesieve-log) of user 74 dumped core.
Stack trace of thread 440934: #0 0x000079c6cfa9890c n/a (libc.so.6 + 0x9890c) #1 0x000079c6cfa3e3a0 raise (libc.so.6 + 0x3e3a0) #2 0x000079c6cfa2557a abort (libc.so.6 + 0x2557a) #3 0x000079c6cfe5ffff n/a (libdovecot.so.0 + 0x5ffff) #4 0x000079c6cff4f095 n/a (libdovecot.so.0 + 0x14f095) #5 0x000079c6cfe5feb6 i_panic (libdovecot.so.0 + 0x5feb6) #6 0x000062e017df80de n/a (/usr/lib/dovecot/managesieve-login + 0x30de) #7 0x000062e017df8887 n/a (/usr/lib/dovecot/managesieve-login + 0x3887) #8 0x000079c6cff681f4 io_loop_call_io (libdovecot.so.0 + 0x1681f4) #9 0x000079c6cff6a2db io_loop_handler_run_internal (libdovecot.so.0 + 0x16a2db) #10 0x000079c6cff6a397 io_loop_handler_run (libdovecot.so.0 + 0x16a397) #11 0x000079c6cff6a599 io_loop_run (libdovecot.so.0 + 0x16a599) #12 0x000079c6cfeabce1 master_service_run (libdovecot.so.0 + 0xabce1) #13 0x000079c6d02f4350 login_binary_run (libdovecot-login.so.0 + 0x14350) #14 0x000079c6cfa27635 n/a (libc.so.6 + 0x27635) #15 0x000079c6cfa276e9 __libc_start_main (libc.so.6 + 0x276e9) #16 0x000062e017df8405 n/a (/usr/lib/dovecot/managesieve-login + 0x3405) ELF object binary architecture: AMD x86-64 Nov 25 14:21:08 mail.schafweide.org dovecot[440904]: managesieve-login: Fatal: master: service(managesieve-login): child 440934 killed with signal 6 (core dumped)
I'm using dovecot/pigeonhole 2.4.2 on Arch.
Regards Bjoern
dovecot mailing list -- dovecot@dovecot.org To unsubscribe send an email to dovecot-leave@dovecot.org
dovecot mailing list -- dovecot@dovecot.org To unsubscribe send an email to dovecot-leave@dovecot.org
Hi,
Oh, and you need to (temporarily!) run managesieve-login without chroot as well to get a proper backtrace.
I hope this is helpful:
0x00007ffff7b256cb in io_stream_ssl_global_init (set=0x55555557d348, error_r=0x7fffffffe168) at ../lib-ssl-iostream/iostream-ssl.c:55 55 return ssl_vfuncs->global_init(set, error_r); (gdb) thread apply all backtrace full
Thread 1 (Thread 0x7ffff7ee41c0 (LWP 626847) "managesieve-log"): #0 0x00007ffff7b256cb in io_stream_ssl_global_init (set=0x55555557d348, error_r=0x7fffffffe168) at ../lib-ssl-iostream/iostream-ssl.c:55 No locals. #1 0x00007ffff7f9204b in login_ssl_init () at /usr/src/debug/dovecot/dovecot-2.4.2/src/login-common/main.c:343 ssl_set = 0x55555557d348 error = 0x200000 <error: Cannot access memory at address 0x200000> #2 main_preinit () at /usr/src/debug/dovecot/dovecot-2.4.2/src/login-common/main.c:355 max_fds = <optimized out> __func__ = <optimized out> #3 login_binary_run (binary=<optimized out>, argc=<optimized out>, argv=<optimized out>) at /usr/src/debug/dovecot/dovecot-2.4.2/src/login-common/main.c:528 service_flags = <optimized out> login_socket = 0x7ffff7f96b66 "login" error = 0x1000000 <error: Cannot access memory at address 0x1000000> c = <optimized out> input = {config_path = 0x0, config_fd = 0, preserve_environment = false, preserve_user = false, preserve_home = false, no_service_filter = false, check_full_config = false, hide_obsolete_warnings = false, no_protocol_filter = false, no_key_validation = false, reload_config = false, never_exec = false, always_exec = false, return_config_fd = false, use_sysexits = false, protocol = 0x55555555b349 "sieve"} output = {specific_protocols = 0x55555555f110, config_fd = -1, permission_denied = false} #4 0x00007ffff7627635 in __libc_start_call_main (main=main@entry=0x5555555573c0 <main>, argc=argc@entry=1, argv=argv@entry=0x7fffffffe348) at ../sysdeps/nptl/libc_start_call_main.h:58 self = <optimized out> result = <optimized out> unwind_buf = {cancel_jmp_buf = {{jmp_buf = {0, -2527813956672304006, 140737488347976, 1, 140737354125312, 93824992270392, -2527813956649235334, -2527795227094430598}, mask_was_saved = 0}}, priv = {pad = {0x0, 0x0, 0x0, 0x7fffffffe348}, data = {prev = 0x0, cleanup = 0x0, canceltype = 0}}} not_first_call = <optimized out> #5 0x00007ffff76276e9 in __libc_start_main_impl (main=0x5555555573c0 <main>, argc=1, argv=0x7fffffffe348, init=<optimized out>, fini=<optimized out>, rtld_fini=<optimized out>, stack_end=0x7fffffffe338) at ../csu/libc-start.c:360 No locals. #6 0x0000555555557405 in _start ()
Regards Bjoern
Hi!
You can work it around by adding --authmech login
sieve-connect -s HOST -p 4190 -u USER --remotesieve NAME --localsieve /path/to/file.sieve --download --authmech login
What exact command line did you use for sieve-connect?
Op 25-11-2025 om 14:51 schreef Bjoern Franke via dovecot:
Hi,
while trying to connect with "sieve-connect", managesieve crashed:
Nov 25 14:21:08 mail.schafweide.org dovecot[440904]: managesieve-login: Panic: file client.c: line 350 (managesieve_client_input_next_cmd): assertion failed: (args != NULL) Nov 25 14:21:08 mail.schafweide.org systemd-coredump[440938]: [🡕] Process 440934 (managesieve-log) of user 74 dumped core.
Stack trace of thread 440934: #0 0x000079c6cfa9890c n/a (libc.so.6 + 0x9890c) #1 0x000079c6cfa3e3a0 raise (libc.so.6 + 0x3e3a0) #2 0x000079c6cfa2557a abort (libc.so.6 + 0x2557a) #3 0x000079c6cfe5ffff n/a (libdovecot.so.0 + 0x5ffff) #4 0x000079c6cff4f095 n/a (libdovecot.so.0 + 0x14f095) #5 0x000079c6cfe5feb6 i_panic (libdovecot.so.0 + 0x5feb6) #6 0x000062e017df80de n/a (/usr/lib/dovecot/managesieve-login + 0x30de) #7 0x000062e017df8887 n/a (/usr/lib/dovecot/managesieve-login + 0x3887) #8 0x000079c6cff681f4 io_loop_call_io (libdovecot.so.0 + 0x1681f4) #9 0x000079c6cff6a2db io_loop_handler_run_internal (libdovecot.so.0 + 0x16a2db) #10 0x000079c6cff6a397 io_loop_handler_run (libdovecot.so.0 + 0x16a397) #11 0x000079c6cff6a599 io_loop_run (libdovecot.so.0 + 0x16a599) #12 0x000079c6cfeabce1 master_service_run (libdovecot.so.0 + 0xabce1) #13 0x000079c6d02f4350 login_binary_run (libdovecot-login.so.0 + 0x14350) #14 0x000079c6cfa27635 n/a (libc.so.6 + 0x27635) #15 0x000079c6cfa276e9 __libc_start_main (libc.so.6 + 0x276e9) #16 0x000062e017df8405 n/a (/usr/lib/dovecot/managesieve-login + 0x3405) ELF object binary architecture: AMD x86-64 Nov 25 14:21:08 mail.schafweide.org dovecot[440904]: managesieve-login: Fatal: master: service(managesieve-login): child 440934 killed with signal 6 (core dumped)
I'm using dovecot/pigeonhole 2.4.2 on Arch.
Regards Bjoern
dovecot mailing list -- dovecot@dovecot.org To unsubscribe send an email to dovecot-leave@dovecot.org
On 11/29/25 4:33 AM, Stephan Bosch via dovecot wrote:
What exact command line did you use for sieve-connect? I mentioned it in my workaround, just replace login with plain.
sieve-connect -s HOST -p 4190 -u USER --remotesieve SCRIPT --localsieve /tmp/script.sieve --download --authmech plain
it asserts in case of plain auth but login auth works.
participants (3)
-
Bjoern Franke
-
Sergej Pupykin
-
Stephan Bosch