On 2/20/20 2:05 PM, Simon Colston wrote:
I'm running podman version 1.8.0 on Fedora 31
If I run with this command
podman run --userns=keep-id \ --name=mysql \ --env="MYSQL_ROOT_PASSWORD=mysql" --publish 3306:3306 \ --volume=/home/simon/servers/mysql/var/lib/mysql:/var/lib/mysql:Z mysql:8.0
I can connect with the mysql client with no errors.
mysql -u root -p -h 127.0.0.1 -P 3306 Enter password: Welcome to the MySQL monitor. Commands end with ; or \g. etc.
If I then do:
podman stop mysql podman start mysql
then I can no longer connect with the mysql client.
mysql -u root -p -h 127.0.0.1 -P 3306 Enter password: ERROR 2003 (HY000): Can't connect to MySQL server on '127.0.0.1' (111)
I then reset everything: podman stop mysql podman rm mysql and delete all the database files.
If I remove the --userns=keep-id argument I can connect after a 'stop' and a 'start'. (Except that the database files now have an odd user id that I can only delete as root)
So, --userns=keep-id is the only difference and causing some sort of problem after a restart.
I had this working using podman version 1.7 and I think that (and maybe the kernel) are the only things that have changed.
Am I doing something wrong? Have I found a bug and would you like me to submit it?
Simon
_______________________________________________ Podman mailing list -- podman@lists.podman.io To unsubscribe send an email to podman-leave@lists.podman.io
That looks like something is destroying and not recreating the iptables rules. Matt do you have any ideas?