Bug 1370990

Summary: [abrt] festival: EST_Item::EST_Item(): festival killed by SIGSEGV
Product: [Fedora] Fedora Reporter: robert fairbrother <noobusinghacks>
Component: festivalAssignee: Bruno Wolff III <bruno>
Status: CLOSED WONTFIX QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 24CC: bruno, bugzilla.redhat, edouard, igiwatson, mattdm, rebus, zlynx
Target Milestone: ---   
Target Release: ---   
Hardware: i686   
OS: Unspecified   
URL: https://retrace.fedoraproject.org/faf/reports/bthash/997b0a358f5ccfdd8daf8b5d32e6dff1c4f02cce
Whiteboard: abrt_hash:bddc39d49cc325258f4669b22abb2c4661a9b1cb;VARIANT_ID=workstation;
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2016-10-19 21:22:43 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
File: backtrace
none
File: cgroup
none
File: core_backtrace
none
File: dso_list
none
File: environ
none
File: exploitable
none
File: limits
none
File: maps
none
File: mountinfo
none
File: namespaces
none
File: open_fds
none
File: proc_pid_status
none
File: var_log_messages none

Description robert fairbrother 2016-08-29 06:10:36 UTC
Description of problem:
festival had a segmentation fault when i run from the gnome-terminal may be related to other bugs

Version-Release number of selected component:
festival-1.96-32.fc24

Additional info:
reporter:       libreport-2.7.2
backtrace_rating: 4
cmdline:        festival
crash_function: EST_Item::EST_Item
executable:     /usr/bin/festival
global_pid:     3517
kernel:         4.6.7-300.fc24.i686
pkg_fingerprint: 73BD E983 81B4 6521
pkg_vendor:     Fedora Project
runlevel:       N 5
type:           CCpp
uid:            1000

Truncated backtrace:
Thread no. 1 (10 frames)
 #0 EST_Item::EST_Item at EST_Item.cc:167
 #1 EST_Relation::append at EST_Relation.cc:98
 #3 add_token at text_aux.cc:47
 #4 FT_Text_Utt at text.cc:92
 #5 leval at slib.cc:1383
 #6 leval_progn at slib_core.cc:81
 #7 leval at slib.cc:1418
 #8 leval_progn at slib_core.cc:81
 #9 leval at slib.cc:1418
 #10 leval_args at slib.cc:1305

Comment 1 robert fairbrother 2016-08-29 06:10:41 UTC
Created attachment 1195141 [details]
File: backtrace

Comment 2 robert fairbrother 2016-08-29 06:10:42 UTC
Created attachment 1195142 [details]
File: cgroup

Comment 3 robert fairbrother 2016-08-29 06:10:43 UTC
Created attachment 1195143 [details]
File: core_backtrace

Comment 4 robert fairbrother 2016-08-29 06:10:44 UTC
Created attachment 1195144 [details]
File: dso_list

Comment 5 robert fairbrother 2016-08-29 06:10:45 UTC
Created attachment 1195145 [details]
File: environ

Comment 6 robert fairbrother 2016-08-29 06:10:47 UTC
Created attachment 1195146 [details]
File: exploitable

Comment 7 robert fairbrother 2016-08-29 06:10:48 UTC
Created attachment 1195147 [details]
File: limits

Comment 8 robert fairbrother 2016-08-29 06:10:49 UTC
Created attachment 1195148 [details]
File: maps

Comment 9 robert fairbrother 2016-08-29 06:10:50 UTC
Created attachment 1195149 [details]
File: mountinfo

Comment 10 robert fairbrother 2016-08-29 06:10:51 UTC
Created attachment 1195150 [details]
File: namespaces

Comment 11 robert fairbrother 2016-08-29 06:10:52 UTC
Created attachment 1195151 [details]
File: open_fds

Comment 12 robert fairbrother 2016-08-29 06:10:53 UTC
Created attachment 1195152 [details]
File: proc_pid_status

Comment 13 robert fairbrother 2016-08-29 06:10:55 UTC
Created attachment 1195153 [details]
File: var_log_messages

Comment 14 robert fairbrother 2016-09-06 12:24:05 UTC
Similar problem has been detected:

there were a number of missing builddep for festival there mostly texlive packages i tryed running it again but i still get segmnetation faults

reporter:       libreport-2.7.2
backtrace_rating: 4
cmdline:        festival
crash_function: EST_Item::EST_Item
executable:     /usr/bin/festival
global_pid:     3616
kernel:         4.6.7-300.fc24.i686
package:        festival-1.96-32.fc24
pkg_fingerprint: 73BD E983 81B4 6521
pkg_vendor:     Fedora Project
reason:         festival killed by SIGSEGV
runlevel:       N 5
type:           CCpp
uid:            1000

Comment 15 Matthew Miller 2016-10-19 21:22:43 UTC
The Festival package in Fedora needs a lot of work to update to a newer release, fix a lot of bugs, and generally clean up the packaging. And, despite best intentions, we just haven't had the developer interest in doing so. 

The current plan is to retire Festival from Fedora. See thread at https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/message/UOL4ETKKOFGTFZZ36V726OF7UTHYMEYP/ for discussion.

Despite being large, slow, and fragile, Festival is interesting software and as far as I know can produce the best quality results of any open-source TTS system. It would be nice if there is interest in continuing it, but that should probably be done from a clean slate in any case.

In the meantime, I think it's most honest to close the currently-open bugs as "WONTFIX". Thanks everyone for your reports and effort in making Fedora better.

Comment 16 Michal Ambroz 2016-11-06 02:57:22 UTC
Hi,
I have tried to rebuild festival with version 2.4 (latest release). Add for the test oald, add patch from Debian to be more strict about boundaries in HTS and anable support for HTS2.1 format. 

Nothing helped so far. The bug remains exactly the same.

Here is the 2.4 src.rpm and a backtrace for the case somebody would like to start where I ended:
https://rebus.fedorapeople.org/SRPMS/festival-2.4-1.fc24.src.rpm


$ gdb --args festival --tts
run
Hello World

gdb$ bt
#0  EST_Item::EST_Item (this=0x62a370, rel=0x62a3b0, li=0x0) at EST_Item.cc:167
#1  0x00007ffff742fefd in EST_Relation::append (this=0x62a3b0, si=0x0) at EST_Relation.cc:98
#2  0x00007ffff7b62f3d in add_token (u=u@entry=0x700590, t=...) at text_aux.cc:47
#3  0x00007ffff7b62393 in tts_chunk_stream (ts=..., app_tok=app_tok@entry=0x7ffff7b61840 <tts_raw_token(EST_Item*)>, app_utt=app_utt@entry=0x7ffff7b61850 <tts_raw_utt(LISP)>, eou_tree=0x7fffe2f135b0, utt=0x7fffe306af30, utt@entry=0x0) at text.cc:242
#4  0x00007ffff7b62923 in tts_file_raw (filename=filename@entry=0x7fffe3068530) at text.cc:169
#5  0x00007ffff7b62ba7 in tts_file (filename=filename@entry=0x7fffe3068530, mode=0x7fffe2f23230) at text.cc:106
#6  0x00007ffff782c162 in leval (x=<optimized out>, qenv=qenv@entry=0x0) at slib.cc:1432
#7  0x00007ffff7b26ddc in festival_eval_command (command=...) at festival.cc:187
#8  0x00007ffff7b27612 in festival_say_file (fname=...) at festival.cc:243
#9  0x000000000040284f in ?? ()
#10 0x00000000004017e9 in ?? ()
#11 0x00007ffff5e9f731 in __libc_start_main () from /lib64/libc.so.6
#12 0x0000000000401849 in ?? ()

Comment 17 Edouard Duliège 2016-11-06 13:39:58 UTC
Similar problem has been detected:

text file with "this is an example"

try festival --tts example.txt

segfault everytime.

reporter:       libreport-2.7.2
backtrace_rating: 3
cmdline:        festival --tts
crash_function: EST_Item::EST_Item
executable:     /usr/bin/festival
global_pid:     3210
kernel:         4.8.4-200.fc24.x86_64
package:        festival-1.96-32.fc24
pkg_fingerprint: 73BD E983 81B4 6521
pkg_vendor:     Fedora Project
reason:         festival killed by SIGSEGV
runlevel:       N 5
type:           CCpp
uid:            1000

Comment 18 Michal Ambroz 2016-11-10 17:25:54 UTC
Yes Eduard both "--tts" from file or stdin fail within the same functionality.
Same also for:

$ festival

Festival Speech Synthesis System 2.4:release December 2014
Copyright (C) University of Edinburgh, 1996-2010. All rights reserved.

clunits: Copyright (C) University of Edinburgh and CMU 1997-2010
clustergen_engine: Copyright (C) Carnegie Mellon University 2005-2014
hts_engine: All rights reserved.
For details type `(festival_warranty)'
festival> (SayText 'hello)
Segmentation fault (core dumped)

It triggers the same bug I believe ... always EST_Item::EST_Item.
Problem is that it is not hit on the first and every invocation of this class/method. Just at some point down the street it crashes - probably some memory management.