Date: Tue, 27 Aug 2024 00:33:58 +0000 From: bugzilla-noreply@freebsd.org To: ports-bugs@FreeBSD.org Subject: [Bug 259795] databases/galera26: cannot initialize new cluster Message-ID: <bug-259795-7788-1PYXUk4sLt@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-259795-7788@https.bugs.freebsd.org/bugzilla/> References: <bug-259795-7788@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D259795 Mike Andrews <mandrews@bit0.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |mandrews@bit0.com --- Comment #5 from Mike Andrews <mandrews@bit0.com> --- I know this ticket is almost 3 years old, but I'm replying just so the next person that finds this via Google has another thing to try: I've just run into this exact same problem this past weekend, trying to set= up a test cluster on some test VMs, using FreeBSD 14.1-p3 and various MariaDB versions: I tried all of 11.4, 10.11, 10.6, and even 10.5. All of them gav= e me the same FD_CLOEXEC, Bad file descriptor error when trying to bootstrap the cluster from scratch, followed by having to kill -9 mariadbd to get it to q= uit. The solution in my case was absolutely bizarre: Change the VM from one CPU core to two CPU cores. Yes, seriously. All I can figure is there's some kind of race condition, scheduling issue, = or something else similar. This weird insight came about when I tried doing the same with an identical my.cnf, identical MariaDB install, and identical FreeBSD install, but on different bare metal hardware, and it worked perfectly there. From there it went to comparing differences between the bare metal hardware and the hypervisor hardware and I just stumbled awkwardly into that solution from there. Glitches like this are why I drink. :) --=20 You are receiving this mail because: You are the assignee for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-259795-7788-1PYXUk4sLt>