Arch Linux segfault in config

[kaadmy]

I'm using the latest version of Arch Linux (as of 2017-05-28, 10:22 PDT), and i3status segfaults every time I run it.
I've narrowed it down to comments that don't have any text on them in specific situations, examples:

Works:

# A
#

# Dummy

# Dummy
# Comment here
#

#Dummy
# Comment here
#

#
general {
}

Doesn't work:

#

#
#

#
# A

#
# Comment here
#

Original issue:

I'm using the latest version of Arch Linux (as of 2017-05-28, 10:22 PDT), and i3status segfaults every time I run it.

I've narrowed it down to the ```general``` block in my config (it's in `~/.config/i3status/config`)
If the block is there at all (even empty), it segfaults.

**Edit:** removing the `general` block no longer fixes it, will do more testing

[Airblader]

Is this the same as #224?

[kaadmy]

I don't think so, I ran the linked config in that issue and it works fine (minus it complaining about the colors not being in a string)

[jbernhard]

This is related to a recent update to libconfuse 3.1. On Arch, the problem can be worked around by downgrading i3status 2.11-2 -> 2.11-1 and confuse 3.1-1 -> 3.0-1. I'm not yet sure what the specific issue is.

[kaadmy]

I just patched my config since other that this issue it still works fully.

[oberon-manjaro]

Manjaro uses a custom build of i3status and it simply needed a rebuilt against confuse 3.1. Since then it is working normally again.

[Airblader]

Thanks!

[oberon-manjaro]

Well, I'm afraid we have two different issues here. Even after the rebuild certain config causes confuse related segfault. I didn't realize it first because my config is still working.
On the Manjaro forum someone just posted the attached offending config
i3status.txt

(gdb) run
Starting program: /usr/bin/i3status 
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/usr/lib/libthread_db.so.1".

Program received signal SIGSEGV, Segmentation fault.
0x00007f6c787485da in ?? () from /usr/lib/libconfuse.so.1
(gdb) bt
#0  0x00007f6c787485da in ?? () from /usr/lib/libconfuse.so.1
#1  0x00007f6c787493fb in cfg_yylex () from /usr/lib/libconfuse.so.1
#2  0x00007f6c78746ac8 in ?? () from /usr/lib/libconfuse.so.1
#3  0x00007f6c78747359 in cfg_parse_fp () from /usr/lib/libconfuse.so.1
#4  0x00007f6c78747428 in cfg_parse () from /usr/lib/libconfuse.so.1
#5  0x00000000004057d9 in main ()
(gdb)

[oberon-manjaro]

Ok. This is funny. In the just attached config simply removing the commented bit at the beginning fixes it! 😜

[oberon-manjaro]

In fact, just removing the first line full of hashes is enough.

[HiTuX]

you are right @Oberon2007.
only one # could cause the segfault peoblem.
I think, the first line should be:

• empty
• # followed by anything other then a #
• or just the first general statement

[oberon-manjaro]

#a
general {
...

^ works

#
general {
...

^ fails

##
general {
...

^ fails

#####a
general {
...

^ works
😆

[oberon-manjaro]

Looks very much like a bug in latest liibconfuse. I've opened an issue there.

[HiTuX]

for me:

general {
...

Works

[troglobit]

Yep, was a regression in libConfuse introduced in v3.1. Fixed in libconfuse/libconfuse@7d70dd2 -- a new release will be out at the latest this weekend.

[oberon-manjaro]

Awesome! Thank you!

[troglobit]

libConfuse v2.3 released now ... https://github.com/martinh/libconfuse/releases/tag/v3.2