[mythtv-users] mythtranscode 0.32 segfaults

Calvin Dodge caldodge at gmail.com
Sat Jun 18 14:11:34 UTC 2022


I grabbed the latest update from the rpmfusion site.

The good news: I can now ask Mythtv to transcode files
The bad new:  mythtranscode consistently segfaults after a few seconds

This is related to 0.32, since I tried it manually when that Mythtv
bug prevented me from requesting it via a frontend.


Here's the last bit of an strace.  Note that it crashes shortly after
opening the output file.

t_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
futex(0x560bfa838220, FUTEX_WAIT_BITSET_PRIVATE|FUTEX_CLOCK_REALTIME,
0, NULL, FUTEX_BITSET_MATCH_ANY) = 0
futex(0x560bfa8381d0, FUTEX_WAKE_PRIVATE, 1) = 0
openat(AT_FDCWD, "/myth/tv/4764_20220601000000.mpg", O_RDONLY) = 21
fcntl(21, F_SETFD, FD_CLOEXEC)          = 0
newfstatat(21, "", {st_mode=S_IFREG|0664, st_size=5686808576, ...},
AT_EMPTY_PATH) = 0
lseek(21, 0, SEEK_SET)                  = 0
read(21, "\0\0\1\272D\0\4\0\4\1CC[\370\0\0\1\273\0\f\241\241\255\4!\377\340\340\346\300\300
"..., 32768) = 32768
read(21, "\0\0\1\272D\0\4\1\204\1CC[\370\0\0\1\340\7\354\200\0\1\377I\355@\33\376d4]"...,
32768) = 32768
read(21, "\0\0\1\272D\0\4\3\4\1CC[\370\0\0\1\340\7\354\200\0\1\377\0s\0\r\302\3414\4"...,
32768) = 32768
read(21, "\0\0\1\272D\0\4\4\204\1CC[\370\0\0\1\340\7\354\200\0\1\377\245\347\255\325sU\345\265"...,
32768) = 32768
read(21, "\0\0\1\272D\0\4\6\4\1CC[\370\0\0\1\340\7\354\200\0\1\377;\10\323\206\200\341\"$"...,
32768) = 32768
newfstatat(21, "", {st_mode=S_IFREG|0664, st_size=5686808576, ...},
AT_EMPTY_PATH) = 0
newfstatat(21, "", {st_mode=S_IFREG|0664, st_size=5686808576, ...},
AT_EMPTY_PATH) = 0
lseek(21, 5686558576, SEEK_SET)         = 5686558576
read(21, "\242\226)\347\216Ycr\26\325y\226\36]\230\34v'\245\246)'\222i\347\222y%\226\211(\222"...,
32768) = 32768
read(21, "D2\237\257\23{\234n\273I\3006&\206\360\20\342\32\313\343\t\250O\31\2678\343\257<\5\10I"...,
32768) = 32768
read(21, "Ig,\264\34k\343Q\354\200\260\305s-8\321\253:\373\0\4\240\32\6\0`\0\324\232B\fI"...,
32768) = 32768
read(21, ")H\213\224\245\".R\224\210\271JR\"\345)H\213\224\245\".R\224\210\271JR\"\345)H"...,
32768) = 32768
read(21, "\7\36\337\353R;\241\306\224\5.\33*\216\351=|\267d\17\316y\253\234\325\5\3145\234p\215\221"...,
32768) = 32768
read(21, "\331'\203\10\0\300\3\232\300\0\177\3x\0\306p\0\0\1+*p\20\7\232\3rl\3\2116\4"...,
32768) = 32768
read(21, "\0\327\0\210\210n\3\10\0\327\08Y\2\270\6\216\23\310\315\33\240\34\v\256\200\374<\3\252\314\0"...,
32768) = 32768
read(21, "\18\fHh\2\210%\6#\375\231\16\354\252yhOG\346U\212\350|>\375\260
\1`\6\205"..., 32768) = 20624
read(21, "", 32768)                     = 0
read(21, "", 32768)                     = 0
read(21, "", 32768)                     = 0
newfstatat(21, "", {st_mode=S_IFREG|0664, st_size=5686808576, ...},
AT_EMPTY_PATH) = 0
lseek(21, 0, SEEK_SET)                  = 0
newfstatat(21, "", {st_mode=S_IFREG|0664, st_size=5686808576, ...},
AT_EMPTY_PATH) = 0
--- SIGSEGV {si_signo=SIGSEGV, si_code=SEGV_MAPERR, si_addr=0x40} ---
write(6, "\v\0\0\0\1\0\0\0@\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0", 24) = 24
rt_sigaction(SIGSEGV, {sa_handler=SIG_DFL, sa_mask=[SEGV],
sa_flags=SA_RESTORER|SA_RESTART, sa_restorer=0x7facc9a88a70},
{sa_handler=0x7facca88ae60, sa_mask=[],
sa_flags=SA_RESTORER|SA_ONSTACK|SA_RESTART|SA_SIGINFO,
sa_restorer=0x7facc9a88a70}, 8) = 0
clock_nanosleep(CLOCK_REALTIME, 0, {tv_sec=1, tv_nsec=0}, 0x560bfa7353b0) = 0
write(2, "Handling Segmentation fault\n", 28) = 28
gettid()                                = 604018
getpid()                                = 604018
tgkill(604018, 604018, SIGSEGV)         = 0
rt_sigreturn({mask=[]})                 = 64
--- SIGSEGV {si_signo=SIGSEGV, si_code=SI_TKILL, si_pid=604018, si_uid=1000} ---
+++ killed by SIGSEGV (core dumped) +++


The system is running Fedora 36 on a Ryzen 5 5600G.
This is rpm information:

Name        : mythtv-common
Version     : 32.0
Release     : 1.36.20220605git7077a824d2.fc36

Does anyone have a clue as to what's happening?


More information about the mythtv-users mailing list