Messed up update over SSH; No libcursesw.so; How to recover

adityam · 2015-10-11 01:00:04

Hi,

I was updating my system remotely via ssh using

$sudo pacman -Syu

All the packages were downloaded correctly but the ssh connection timed out when only a fraction of packages had been installed (let's not get into the details of why the connection time out; most likely because my laptop hibernated).

Now when I try to ssh, I get

zsh: error while loading shared libraries: libncursesw.so.5: cannot open shared object file: No such file or directory
Connection to <hostname> closed.

It appears that libcursesw was updated but zsh (my default shell) was not. So, what can I do now.

I won't have physical access to the server until three days. I was locally logged on to the server, so if the current desktop environment is still running, it may be possible for me to recover by running pacman -u.

However, in the meantime, is there anything that I can try remotely?

Thanks,
Adi

Trilby · 2015-10-11 01:19:53

ssh and run pacman rather than a shell:

ssh user@host /usr/bin/pacman -Syu

Although, depending on what actually happened pacman may not work either.

In the future you may want to consider using screen or tmux within ssh sessions so a random disconnect will not interrupt important operations.

adityam · 2015-10-11 03:06:46

Trilby wrote:

ssh and run pacman rather than a shell:
ssh user@host /usr/bin/pacman -Syu
Although, depending on what actually happened pacman may not work either.

It doesn't work. I get the same error message:

zsh: error while loading shared libraries: libncursesw.so.5: cannot open shared object file: No such file or directory

bebopson · 2018-08-25 11:27:18

I had also had the connection drop, but was able to log back in. At the time the connection dropped, pacman was waiting for me to confirm new key installation. I tried to run pacman again but it warned me:

error: could not lock database: File exists
if you're sure a package manager is not already
running, you can remove /var/lib/pacman/db.lck

I ran ps aux | grep pacman which revealed the pre-connection-drop pacman process is still running.

root 1290 0.0 0.0 84096 5628 pts/0 S+ 06:12 0:00 sudo pacman -Syu
root 1291 0.7 0.6 227520 47172 pts/0 S+ 06:12 0:27 pacman -Syu
root 2272 0.0 0.2 147144 22232 ? Ss 06:35 0:00 dirmngr --daemon --homedir /etc/pacman.d/gnupg

What's the appropriate thing to do in this situation?

WorMzy · 2018-08-25 13:58:26

bebopson wrote:

What's the appropriate thing to do in this situation?

Create a new topic for your issue, and don't necrobump.

https://wiki.archlinux.org/index.php/Co … bumping.22

Closing.

Arch Linux

#1 2015-10-11 01:00:04

Messed up update over SSH; No libcursesw.so; How to recover

#2 2015-10-11 01:19:53

Re: Messed up update over SSH; No libcursesw.so; How to recover

#3 2015-10-11 03:06:46

Re: Messed up update over SSH; No libcursesw.so; How to recover

#4 2018-08-25 11:27:18

Re: Messed up update over SSH; No libcursesw.so; How to recover

#5 2018-08-25 13:58:26

Re: Messed up update over SSH; No libcursesw.so; How to recover

Board footer