#988178 nis: ypbind fails to start when already running in a container

Package:
nis
Source:
nis
Submitter:
Brian Morris
Date:
2021-05-07 14:21:02 UTC
Severity:
important
Tags:
#988178#5
Date:
2021-05-07 00:13:35 UTC
From:
To:
Dear Maintainer,

   * What led up to the situation?
     System has several lxd containers that run ypbind. When the system was
     restarted (or even ran dist-upgrade just prior to restart), pybind failed
     to start. I have determined this is because start-stop-daemon is confused
     by processes named ypbind visible in the containers even though they are
     not "local" and not accessible by the host system. I have also determined
     that simply passing "--pidfile /var/run/ypbind.pid" to start-stop-daemon
     -- as is already done for the stop case -- fixes the issue.
   * What exactly did you do (or not do) that was effective (or
     ineffective)? Manually running "ypbind" works, as does the above fix.
#988178#10
Date:
2021-05-07 00:28:24 UTC
From:
To:
I also looked at the source of the newer ypbind-mt package in the Testing branch and it appears to have the same issue.
#988178#15
Date:
2021-05-07 09:02:16 UTC
From:
To:
Hi

I guess you are talking about the init script still provided?
#988178#20
Date:
2021-05-07 14:19:44 UTC
From:
To:
Yes, I have not specifically tested this situation with a newer distro install but I looked at the nis init script which appears to have been moved to the ypbind-mt package, and the invocation of start-stop-daemon for starting ypbind does not pass --pidfile.


Hi

I guess you are talking about the init script still provided?