Date: Mon, 13 Sep 2021 11:48:40 -0700 From: Freddie Cash <fjwcash@gmail.com> To: joe mcguckin <joe@via.net> Cc: freebsd-fs <freebsd-fs@freebsd.org> Subject: Re: more zfs questions Message-ID: <CAOjFWZ43jkF3aOUuUQS6ZCmuxfdwoC6ZT77%2BotWF7i7t39brbA@mail.gmail.com> In-Reply-To: <2BD7455F-56B2-4152-AB15-B597DF3349A3@via.net>
index | next in thread | previous in thread | raw e-mail
On Mon, Sep 13, 2021 at 11:44 AM joe mcguckin <joe@via.net> wrote: > I had to replace a drive. > > - zpool status: the ETA is growing longer even as more of the drive is > being resilvered. Is thst correct?? > This would only happen if the I/O throughput is slowing down over time. > - The activity LEDS corresponding the the VDEV of the failed drive are on > continously. When resilvering, is only the affected VDEV resilvered, or > the entire dataset the VDEV belongs to? > This is normal-ish. When resilvering, data is read from all members of the vdev and written to the new drive. So you should see activity lights for all drives in the vdev, then just the new drive, then all drives in the vdev, then just the new drive, and so on. Depending on fragmentation level, "fullness" of the pool, whether dedupe is enabled, and so on will influence just what pattern the lights take. But it's basically "read from the vdev" / "write to the new disk" until it's done. -- Freddie Cash fjwcash@gmail.comhome | help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAOjFWZ43jkF3aOUuUQS6ZCmuxfdwoC6ZT77%2BotWF7i7t39brbA>