Date: Mon, 18 Mar 2024 02:54:46 +0000 From: bugzilla-noreply@freebsd.org To: fs@FreeBSD.org Subject: [Bug 277389] Reproduceable low memory freeze on 14.0-RELEASE-p5 Message-ID: <bug-277389-3630-w7Gn6RAQOC@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-277389-3630@https.bugs.freebsd.org/bugzilla/> References: <bug-277389-3630@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=3D277389 --- Comment #27 from Mark Millard <marklmi26-fbsd@yahoo.com> --- (In reply to pascal.guitierrez from comment #26) I'll note that I never used mount_nullfs explicitly at any point but I still got the huge difference in VNODE results in "vmstat -z" when doing your test iozone runs, all based just on the: nullfs_load=3D"YES" vfs.nullfs.cache_vnodes=3D0 vs. not distinction. There may be implicit/automatic use involved? ( zfs_load=3D"YES" both ways. ) Because of your hangup status, doing a "vmstat -z" and saving the output before doing the read iozone test would be good. Both with and without the 3 lines for comparison/contrast: the vmstat -z output differences should indicate if internally any difference is being made. I made the suggestion because I've no way to tell if what I've been reporting contributes to your context or not: I've never seen the OOM activity or hangups. But the huge Wired figures that even survive the ARC shrinking suggest potential contribution to an OOM status. --=20 You are receiving this mail because: You are on the CC list for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-277389-3630-w7Gn6RAQOC>