zsh crash after conda activate
etern opened this issue · 2 comments
Describe the bug
only happens with ohmyzsh
- dmesg:
[24263745.738211] zsh[91509]: segfault at 0 ip 00007facd19018c1 sp 00007ffc300bc418 error 4 in libc-2.17.so[7facd1792000+1c4000]
- coredump backtrace:
(gdb) bt
#0 0x00007fb1ed0028c1 in __strlen_sse2_pminub () from /lib64/libc.so.6
#1 0x000000000047d010 in bicat ()
#2 0x000000000043ddd7 in lockhistfile ()
#3 0x000000000043ecc2 in hend ()
#4 0x0000000000440c6f in loop ()
#5 0x000000000044426e in zsh_main ()
#6 0x00007fb1eceb5555 in __libc_start_main () from /lib64/libc.so.6
#7 0x000000000040ee4e in _start ()
Steps to reproduce
- conda activate py39 // nothing wrong here, py39 is one of my conda env
- press Enter // crash here, all my conda envs except
base
env
Expected behavior
not crash
Screenshots and recordings
No response
OS / Linux distribution
CentOS Linux release 7.5.1804
Zsh version
zsh 5.0.2 (x86_64-redhat-linux-gnu)
Oh My Zsh version
master 803e1a7
Terminal emulator
iterm2, ssh to remote server
If using WSL on Windows, which version of WSL
None
Additional context
No response
Hi! The backtrace was indeed useful. Could you share the output of echo $HOST
?
Besides that, could you try executing zsh -f
, and inside that new shell, do echo $HOST
and reproduce the issue?
Thanks!
Thanks for your response, but I cannot reproduce now.
May be caused by some conda package, I will close the issue.
output of HOST has some private information, it's some thing like abc-xyz-123.company.com
, and after zsh -f
the HOST not changed.