I often fail to run apt-get install on my server. Like that:
$ sudo apt-get install tmux
E: Could not get lock /var/lib/dpkg/lock - open (11: Resource
temporarily unavailable)
E: Unable to lock the administration directory (/var/lib/dpkg/), is
another process using it?
Happens a lot, almost every day. Whenever I try to find out who else is locking it with the lsof command, then I am too late and the lock is already gone. Really very weird.
Is there a trick, how I can record for over 24 hrs who else is locking /var/lib/dpkg/lock?
I think there are multiple solutions.
1: write a simple script, for example while sleep 1; do lsof -n|grep /var/lib/dpkg/lock >>/var/log/dpkglocktmp.log; done. It is simple, but not very beautiful, but as a short-time bugtracking, it is to me okay. You will get the list of the processes in a log file.
Extension: while sleep 1; do lsof -n|grep /var/lib/dpkg/lock; done|tee --append /var/log/dpkglocktmp.log will simultaneously write to the console and into the logfile. If you won't keep your ssh connection alive, you can run this in a screen.
2: Linux kernel has a functionality inotify, which enables processes to get signaled on events happened on files they are watching. There are numerous tools around it, the most known is incron. It enables you to call a script on the case of a file event. Probably there more complex tools as well, targeting exactly the problems similar to yours.
I had most similar problems as an apt update/install asked from me something, I forgot that and later tried to run another apt in another ssh session.