bnorton916 said:
Is there a better way to get to the incremental on Tuesday? Currently, I am having to do three recv's. Worse if it was on Friday.
An important aspect to keep in mind is that whenever you create a snapshot it remains in place until you destroyed it. So assuming that
tank/user/norton is mounted as
/home/norton you could easily retrieve your stuff from Tuesday by doing something like this:
Code:
# cd /home/norton/.zfs/snapshot/incremTuesday
# cp $file /home/user/norton/.
There's no direct need to receive and re-import or re-apply the external backup. And because ZFS uses a pool and shares it's data I'd suggest to try this approach first, it makes quickly retrieving a backup a lot easier. Personally I only keep off-site backups as my last means of defense, only to be used when there's no other option available.
bnorton916 said:
Is there a way to just restore a file from a dataset like tank/wife, rather than restoring the entire tank pool?
See above. Whenever you're importing a filesystem back into your pool you can opt to keep it under a different mount point so that you can then access it.
But this is also the caveat of using 'external' snapshots (in my opinion); you can only access them after you imported the whole file system back into your pool. Which can become an issue if you don't have much free space left.
bnorton916 said:
Is there a better way to do this overall?
Better is of course a very personal aspect.
I tend to make snapshots and keep them for a week, then the snapshot gets destroyed again. The good part here is that only changes get stored, ergo it may cost you a lot less space than you might assume right now.
For example, this is what my backup of
/var/log looks like:
Code:
$ zfs list -rt all zroot/var/log
NAME USED AVAIL REFER MOUNTPOINT
zroot/var/log 558M 111G 550M /var/log
zroot/var/log@071013 910K - 302M -
zroot/var/log@081013 838K - 307M -
zroot/var/log@091013 853K - 310M -
zroot/var/log@101013 874K - 313M -
zroot/var/log@111013 896K - 317M -
zroot/var/log@121013 848K - 321M -
zroot/var/log@131013 1.01M - 324M -
For the record; this is on a rather busy server, most of the space mentioned here is consumed by logfiles from Apache. But this approach also allows me to quickly access anything on a specific day.
Maybe you'll find this useful; I wrote a script to handle the whole snapshot rotation. Although
zfs(8) provides an example where a snapshot is created, the other destroyed and the rest renamed I didn't like that approach at all because when it comes to backups (and filesystems in general) I want to keep the required access to a minimum. So I came up with this:
Code:
#!/bin/sh
## Snapshot.ZFS v1.0
##
## A script which will manage ZFS snapshots on the
## filesystem(s) of your choosing.
### Configuration section.
# ZFS pool to use.
POOL="zroot";
# Filesystem(s) to use.
FS="/home /usr/local /var"
# Retention; how many snapshots should be kept?
RETENTION=7
# Recursive; process a filesystem and all it's children?
RECURSE=yes;
### Script definitions <-> ** Don't change anything below this line! **
CURDAT=$(date "+%d%m%y");
PRVDAT=$(date -v-${RETENTION}d "+%d%m%y");
PROG=$(basename $0);
if [ ${RECURSE} == "yes" ]; then
OPTS="-r";
fi
### Script starts here ###
if [ "$1/" == "/" ]; then
echo "Error: no command specified." > /dev/stderr;
echo > /dev/stderr;
echo "Usage:" > /dev/stderr;
echo "${PROG} y : Manage snapshots." > /dev/stderr;
echo > /dev/stderr;
exit 1;
fi
if [ "$1" == "y" ]; then
# Make & clean snapshot(s)
for a in $FS; do
ZFS=$(zfs list -r ${POOL} | grep -e "${a}$" | cut -d ' ' -f1);
if [ "$ZFS/" == "/" ]; then
echo "${PROG}: Can't process ${a}: not a ZFS filesystem." >/dev/stderr;
else
$(zfs snapshot ${OPTS} ${ZFS}@${CURDAT} > /dev/null 2>&1) || echo "${PROG}: Error creating snapshot ${ZFS}@${CURDAT}" > /dev/stderr
$(zfs destroy ${OPTS} ${ZFS}@${PRVDAT} > /dev/null 2>&1) || echo "${PROG}: Error destroying snapshot ${ZFS}@${PRVDAT}" > /dev/stderr
fi
done;
else
echo "Error: wrong parameter used." > /dev/stderr;
echo > /dev/stderr;
echo "Usage:" > /dev/stderr;
echo "${PROG} y : Manage snapshots." > /dev/stderr;
echo > /dev/stderr;
exit 1;
fi
Edit:
If you plan on trying out the script but don't want to take any risks then my advice is to comment out the line with "$(zfs destroy ${OPTS}". You can do so by placing a
# in front of it. This will make sure that snapshots only get created but not destroyed. I did add some failsaves into the script but obviously it best to be save than sore
Also, when looking at this script again I also noticed that in my environment the user who executes this (
root in my case) needs to have
/bin and
/sbin in its PATH. Normally those should already be present, but I figured I'd mention it anyway.
