Compare commits

..

No commits in common. "master" and "0.8" have entirely different histories.

114 changed files with 1068 additions and 2168 deletions

1
.gitignore vendored
View file

@ -8,6 +8,7 @@ doc/man/*.html
doc/man/*.htm
doc/man/*.texi
doc/man/*.man
test/*
.*.swp
doc/man/*.[0-9]
doc/*.xml

View file

@ -1,12 +0,0 @@
stages:
- test
unit_tests:
stage: test
script:
- make test
shellcheck:
stage: test
script:
- make shellcheck

127
Makefile
View file

@ -22,14 +22,14 @@
#
INSTALL=install
CCOLLECT_SOURCE=ccollect
CCOLLECT_DEST=ccollect
CCOLLECT_SOURCE=ccollect.sh
CCOLLECT_DEST=ccollect.sh
LN=ln -sf
ASCIIDOC=asciidoc
DOCBOOKTOTEXI=docbook2x-texi
DOCBOOKTOMAN=docbook2x-man
XSLTPROC=xsltproc
XSL=/usr/local/share/xsl/docbook/html/docbook.xsl
XSL=/usr/share/xml/docbook/stylesheet/nwalsh/html/docbook.xsl
A2X=a2x
prefix=/usr/packages/ccollect-git
@ -41,7 +41,11 @@ manlink=/usr/local/man/man1
path_dir=/usr/local/bin
path_destination=${path_dir}/${CCOLLECT_DEST}
docs_archive_name=docs.tar
# where to publish
host=localhost
dir=/home/users/nico/privat/computer/net/netzseiten/www.nico.schottelius.org/src/software/ccollect
docdir=${dir}/doc
#
# Asciidoc will be used to generate other formats later
@ -75,7 +79,12 @@ DOCBDOCS = ${DOCS:.text=.docbook}
DOC_ALL = ${HTMLDOCS} ${DBHTMLDOCS} ${TEXIDOCS} ${MANPDOCS} ${PDFDOCS}
TEST_LOG_FILE = /tmp/ccollect/ccollect.log
html: ${HTMLDOCS}
htm: ${DBHTMLDOCS}
info: ${TEXIDOCS}
man: ${MANPDOCS}
pdf: ${PDFDOCS}
documentation: ${DOC_ALL}
#
# End user targets
@ -87,15 +96,6 @@ all:
@echo "info: only generate Texinfo"
@echo "man: only generate manpage{s}"
@echo "install: install ccollect to ${prefix}"
@echo "shellcheck: shellcheck ccollect script"
@echo "test: run unit tests"
html: ${HTMLDOCS}
htm: ${DBHTMLDOCS}
info: ${TEXIDOCS}
man: ${MANPDOCS}
pdf: ${PDFDOCS}
documentation: ${DOC_ALL}
install: install-link install-manlink
@ -116,28 +116,21 @@ install-manlink: install-man
#
# Tools
#
TOOLS2=ccollect_add_source
TOOLS2 += ccollect_analyse_logs
TOOLS=ccollect_add_source.sh \
ccollect_analyse_logs.sh \
ccollect_delete_source.sh \
ccollect_list_intervals.sh \
ccollect_logwrapper.sh \
ccollect_list_intervals.sh
TOOLS=ccollect_add_source \
ccollect_analyse_logs \
ccollect_delete_source \
ccollect_list_intervals \
ccollect_logwrapper \
ccollect_list_intervals
# Stick to posix
TOOLSMAN1 = $(TOOLS:ccollect=doc/man/ccollect)
TOOLSMAN = $(TOOLSMAN1:=.text)
TOOLSMAN1 = $(subst ccollect,doc/man/ccollect,$(TOOLS))
TOOLSMAN = $(subst .sh,.text,$(TOOLSMAN1))
TOOLSFP = $(subst ccollect,tools/ccollect,$(TOOLS))
## FIXME: posix make: shell? =>
#t2: $(TOOLSMAN)
t2:
echo $(TOOLS) - $(TOOLSFP)
echo $(TOOLSMAN)
echo $(TOOLSFP)
echo $(TOOLS) - $(TOOLSMAN) - $(TOOLSFP)
# docbook gets .htm, asciidoc directly .html
@ -173,13 +166,13 @@ t2:
#
# Developer targets
#
pub:
git push
update:
@git push
publish-doc: documentation
@echo "Transferring files to ${host}"
@chmod a+r ${DOCS} ${DOC_ALL}
@tar cf ${docs_archive_name} ${DOCS} ${DOC_ALL}
@echo "Documentation files are in ${docs_archive_name}"
@tar c ${DOCS} ${DOC_ALL} | ssh ${host} "cd ${dir}; tar xv"
#
# Distribution
@ -196,55 +189,19 @@ distclean: clean
#
dist: distclean documentation
/tmp/ccollect:
#test: ccollect.sh documentation
test: ccollect.sh
mkdir -p /tmp/ccollect
shellcheck: ./ccollect
shellcheck -s sh -f gcc -x ./ccollect
test-nico: $(CCOLLECT_SOURCE) /tmp/ccollect
cd ./conf/sources/; for s in *; do CCOLLECT_CONF=../ ../../ccollect daily "$$s"; done
CCOLLECT_CONF=./conf ./ccollect.sh daily from-remote
CCOLLECT_CONF=./conf ./ccollect.sh daily local
CCOLLECT_CONF=./conf ./ccollect.sh daily "local-with&ampersand"
CCOLLECT_CONF=./conf ./ccollect.sh daily source-without-destination
CCOLLECT_CONF=./conf ./ccollect.sh daily "source with spaces and interval"
CCOLLECT_CONF=./conf ./ccollect.sh daily to-remote
CCOLLECT_CONF=./conf ./ccollect.sh daily with_exec
CCOLLECT_CONF=./conf ./ccollect.sh daily very_verbose
touch /tmp/ccollect/$$(ls /tmp/ccollect | head -n1).ccollect-marker
CCOLLECT_CONF=./conf ./ccollect -a daily
touch /tmp/ccollect/$$(ls /tmp/ccollect | head -n1).ccollect-marker
CCOLLECT_CONF=./conf ./ccollect -a -p daily
test-dir-source:
mkdir -p /tmp/ccollect/source
cp -R -f ./* /tmp/ccollect/source
test-dir-destination:
mkdir -p /tmp/ccollect/backup
test-dir-destination-chint:
mkdir -p /tmp/ccollect/backup-chint
test-fixed-intervals: $(CCOLLECT_SOURCE) test-dir-source test-dir-destination test-dir-destination-chint
for s in ./test/conf/sources/*; do \
CCOLLECT_CONF=./test/conf ./ccollect -l ${TEST_LOG_FILE} daily "$$(basename "$$s")"; \
test "$$(ls -1 /tmp/ccollect/backup | wc -l)" -gt "0" || { cat ${TEST_LOG_FILE}; exit 1; }; \
done
CCOLLECT_CONF=./test/conf ./ccollect -l ${TEST_LOG_FILE} -a -v daily
test "$$(ls -1 /tmp/ccollect/backup | wc -l)" -gt "0" || { cat ${TEST_LOG_FILE}; exit 1; }
CCOLLECT_CONF=./test/conf ./ccollect -l ${TEST_LOG_FILE} -a -p daily
test "$$(ls -1 /tmp/ccollect/backup | wc -l)" -gt "0" || { cat ${TEST_LOG_FILE}; exit 1; }
@printf "\nFixed intervals test ended successfully\n"
test-interval-changing: $(CCOLLECT_SOURCE) test-dir-source test-dir-destination-chint
rm -rf /tmp/ccollect/backup-chint/*
test "$$(ls -1 /tmp/ccollect/backup-chint | wc -l)" -eq "0" || { cat ${TEST_LOG_FILE}; exit 1; }
printf "3" > ./test/conf/sources/local-with-interval/intervals/daily
for x in 1 2 3 4 5; do CCOLLECT_CONF=./test/conf ./ccollect -l ${TEST_LOG_FILE} daily local-with-interval; done
test "$$(ls -1 /tmp/ccollect/backup-chint | wc -l)" -eq "3" || { cat ${TEST_LOG_FILE}; exit 1; }
printf "5" > ./test/conf/sources/local-with-interval/intervals/daily
for x in 1 2 3 4 5 6 7; do CCOLLECT_CONF=./test/conf ./ccollect -l ${TEST_LOG_FILE} daily local-with-interval; done
test "$$(ls -1 /tmp/ccollect/backup-chint | wc -l)" -eq "5" || { cat ${TEST_LOG_FILE}; exit 1; }
printf "4" > ./test/conf/sources/local-with-interval/intervals/daily
for x in 1 2 3 4 5 6; do CCOLLECT_CONF=./test/conf ./ccollect -l ${TEST_LOG_FILE} daily local-with-interval; done
test "$$(ls -1 /tmp/ccollect/backup-chint | wc -l)" -eq "4" || { cat ${TEST_LOG_FILE}; exit 1; }
printf "3" > ./test/conf/sources/local-with-interval/intervals/daily
@printf "\nInterval changing test ended successfully\n"
test: test-fixed-intervals test-interval-changing
test -f "${TEST_LOG_FILE}"
@printf "\nTests ended successfully\n"
CCOLLECT_CONF=./conf ./ccollect.sh daily delete_incomplete
CCOLLECT_CONF=./conf ./ccollect.sh daily no-source-must-fail
# for s in $$(ls ./conf/sources); do CCOLLECT_CONF=./conf echo ./ccollect.sh daily $$s; done
# CCOLLECT_CONF=./conf ./ccollect.sh -a daily

2
README
View file

@ -7,7 +7,7 @@ ccollect backups (local or remote) data to local or remote destinations.
You can retrieve the latest version of ccollect at [0].
ccollect was inspired by rsnapshot [1], which has some problems:
- configuration parameters have to be TAB seperated
- configuration parameters has to be TAB seperated
- you can not specify per source exclude lists
- no per source pre/post execution support
- no parallel execution

26
TODO
View file

@ -1,26 +0,0 @@
Hinweis:
Zwei Quellen auf ein Ziel funktioniert nicht, da die beiden
Quellen von den unterschiedlichen Verzeichnissen linken.
Listing:
.../* funktioniert nicht teilweise (abhängig von der Shell?)
Isup-check: optional!
line 318/check no_xxx => correct test?
REMOVE ALL PCMD code
Backup to remote can be done via ssh tunnel!
% remote host: Allow backup host to access our sshd
ssh -R4242:localhost:22 backupserver "ccollect interval backupremotehost"
remove $destination (==ddir)
--------------------------------------------------------------------------------
Remote backups:
ccollect_backup_to $host $remote_port $source $interval

929
ccollect
View file

@ -1,929 +0,0 @@
#!/bin/sh
#
# 2005-2013 Nico Schottelius (nico-ccollect at schottelius.org)
# 2016-2019 Darko Poljak (darko.poljak at gmail.com)
#
# This file is part of ccollect.
#
# ccollect is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# ccollect is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with ccollect. If not, see <http://www.gnu.org/licenses/>.
#
# Initially written for SyGroup (www.sygroup.ch)
# Date: Mon Nov 14 11:45:11 CET 2005
# Error upon expanding unset variables:
set -u
#
# Standard variables (stolen from cconf)
#
__mydir="${0%/*}"
__abs_mydir="$(cd "$__mydir" && pwd -P)"
__myname=${0##*/}
#
# where to find our configuration and temporary file
#
CCOLLECT_CONF="${CCOLLECT_CONF:-/etc/ccollect}"
CSOURCES="${CCOLLECT_CONF}/sources"
CDEFAULTS="${CCOLLECT_CONF}/defaults"
CPREEXEC="${CDEFAULTS}/pre_exec"
CPOSTEXEC="${CDEFAULTS}/post_exec"
CMARKER=".ccollect-marker"
TMP="$(mktemp "/tmp/${__myname}.XXXXXX")"
export TMP
CONTROL_PIPE="/tmp/${__myname}-control-pipe"
VERSION="2.10"
RELEASE="2020-08-26"
HALF_VERSION="ccollect ${VERSION}"
FULL_VERSION="ccollect ${VERSION} (${RELEASE})"
#
# CDATE: how we use it for naming of the archives
# DDATE: how the user should see it in our output (DISPLAY)
#
CDATE="date +%Y%m%d-%H%M"
DDATE="date +%Y-%m-%d-%H:%M:%S"
SDATE="date +%s"
#
# LOCKING: use flock if available, otherwise mkdir
# Locking is done for each source so that only one instance per source
# can run.
#
# Use CCOLLECT_CONF directory for lock files.
# This directory can be set arbitrary so it is writable for user
# executing ccollect.
LOCKDIR="${CCOLLECT_CONF}"
# printf pattern: ccollect_<source>.lock
LOCKFILE_PATTERN="ccollect_%s.lock"
LOCKFD=4
#
# locking functions using flock
#
lock_flock()
{
# $1 = source to backup
# shellcheck disable=SC2059
lockfile="${LOCKDIR}/$(printf "${LOCKFILE_PATTERN}" "$1")"
eval "exec ${LOCKFD}> '${lockfile}'"
flock -n ${LOCKFD} && return 0 || return 1
}
unlock_flock()
{
# $1 = source to backup
# shellcheck disable=SC2059
lockfile="${LOCKDIR}/$(printf "${LOCKFILE_PATTERN}" "$1")"
eval "exec ${LOCKFD}>&-"
rm -f "${lockfile}"
}
#
# locking functions using mkdir (mkdir is atomic)
#
lock_mkdir()
{
# $1 = source to backup
# shellcheck disable=SC2059
lockfile="${LOCKDIR}/$(printf "${LOCKFILE_PATTERN}" "$1")"
mkdir "${lockfile}" && return 0 || return 1
}
unlock_mkdir()
{
# $1 = source to backup
# shellcheck disable=SC2059
lockfile="${LOCKDIR}/$(printf "${LOCKFILE_PATTERN}" "$1")"
rmdir "${lockfile}"
}
#
# determine locking tool: flock or mkdir
#
if command -v flock > /dev/null 2>&1
then
lockf="lock_flock"
unlockf="unlock_flock"
else
lockf="lock_mkdir"
unlockf="unlock_mkdir"
fi
#
# unset values
#
PARALLEL=""
MAX_JOBS=""
USE_ALL=""
LOGFILE=""
SYSLOG=""
# e - only errors, a - all output
LOGLEVEL="a"
LOGONLYERRORS=""
#
# catch signals
#
TRAPFUNC="rm -f \"${TMP}\""
# shellcheck disable=SC2064
trap "${TRAPFUNC}" 1 2 15
#
# Functions
#
# check if we are running interactive or non-interactive
# see: http://www.tldp.org/LDP/abs/html/intandnonint.html
_is_interactive()
{
[ -t 0 ] || [ -p /dev/stdin ]
}
#
# ssh-"feature": we cannot do '... read ...; ssh ...; < file',
# because ssh reads stdin! -n does not work -> does not ask for password
# Also allow deletion for files without the given suffix
#
delete_from_file()
{
file="$1"; shift
suffix="" # It will be set, if deleting incomplete backups.
[ $# -eq 1 ] && suffix="$1" && shift
# dirs for deletion will be moved to this trash dir inside destination dir
# - for fast mv operation
trash="$(mktemp -d ".trash.XXXXXX")"
while read -r to_remove; do
mv "${to_remove}" "${trash}" ||
_exit_err "Moving ${to_remove} to ${trash} failed."
set -- "$@" "${to_remove}"
if [ "${suffix}" ]; then
to_remove_no_suffix="$(echo "${to_remove}" | sed "s/$suffix\$//")"
mv "${to_remove_no_suffix}" "${trash}" ||
_exit_err "Moving ${to_remove_no_suffix} to ${trash} failed."
set -- "$@" "${to_remove_no_suffix}"
fi
done < "${file}"
_techo "Removing $* in ${trash}..."
empty_dir=".empty-dir"
mkdir "${empty_dir}" || _exit_err "Empty directory ${empty_dir} cannot be created."
[ "${VVERBOSE}" ] && echo "Starting: rsync -a --delete ${empty_dir} ${trash}"
# rsync needs ending slash for directory content
rsync -a --delete "${empty_dir}/" "${trash}/" || _exit_err "Removing $* failed."
rmdir "${trash}" || _exit_err "Removing ${trash} directory failed"
rmdir "${empty_dir}" || _exit_err "Removing ${empty_dir} directory failed"
_techo "Removing $* in ${trash} finished."
}
display_version()
{
echo "${FULL_VERSION}"
exit 0
}
usage()
{
cat << eof
${__myname}: [args] <interval name> <sources to backup>
ccollect creates (pseudo) incremental backups
-h, --help: Show this help screen
-a, --all: Backup all sources specified in ${CSOURCES}
-e, --errors: Log only errors
-j [max], --jobs [max] Specifies the number of jobs to run simultaneously.
If max is not specified then parallelise all jobs.
-l FILE, --logfile FILE Log to specified file
-p, --parallel: Parallelise backup processes (deprecated from 2.0)
-s, --syslog: Log to syslog with tag ccollect
-v, --verbose: Be very verbose (uses set -x)
-V, --version: Print version information
This is version ${VERSION} released on ${RELEASE}.
Retrieve latest ccollect at http://www.nico.schottelius.org/software/ccollect/
eof
exit 0
}
# locking functions
lock()
{
"${lockf}" "$@" || _exit_err \
"Only one instance of ${__myname} for source \"$1\" can run at one time."
}
unlock()
{
"${unlockf}" "$@"
}
# time displaying echo
# stdout version
_techo_stdout()
{
echo "$(${DDATE}): $*"
}
# syslog version
_techo_syslog()
{
logger -t ccollect "$@"
}
# specified file version
_techo_file()
{
_techo_stdout "$@" >> "${LOGFILE}"
}
# determine _techo version before parsing options
if _is_interactive
then
_techof="_techo_stdout"
else
_techof="_techo_syslog"
fi
# _techo with determined _techo version
_techo()
{
if [ "${LOGLEVEL}" = "a" ]
then
# name is exported before calling this function
# shellcheck disable=SC2154
set -- ${name:+"[${name}]"} "$@"
"${_techof}" "$@"
fi
}
_techo_err()
{
_techo "Error: $*"
}
_exit_err()
{
_techo_err "$@"
rm -f "${TMP}"
exit 1
}
#
# Parse options
#
while [ "$#" -ge 1 ]; do
case "$1" in
-a|--all)
USE_ALL=1
;;
-p|--parallel)
_techo "Warning: -p, --parallel option is deprecated," \
"use -j, --jobs instead."
PARALLEL=1
MAX_JOBS=""
;;
-j|--jobs)
PARALLEL=1
if [ "$#" -ge 2 ]
then
case "$2" in
-*)
;;
*)
MAX_JOBS=$2
shift
;;
esac
fi
;;
-e|--errors)
LOGONLYERRORS="1"
;;
-l|--logfile)
if [ "$#" -ge 2 ]
then
case "$2" in
-*)
_exit_err "Missing log file"
;;
*)
LOGFILE="$2"
shift
;;
esac
else
_exit_err "Missing log file"
fi
;;
-s|--syslog)
SYSLOG="1"
;;
-v|--verbose)
set -x
;;
-V|--version)
display_version
;;
--)
# ignore the -- itself
shift
break
;;
-h|--help|-*)
usage
;;
*)
break
;;
esac
shift
done
# determine _techo version and logging level after parsing options
if [ "${LOGFILE}" ]
then
_techof="_techo_file"
LOGLEVEL="a"
elif _is_interactive
then
if [ "${SYSLOG}" ]
then
_techof="_techo_syslog"
LOGLEVEL="a"
else
_techof="_techo_stdout"
LOGLEVEL="e"
fi
else
_techof="_techo_syslog"
LOGLEVEL="a"
fi
if [ "${LOGFILE}" ] || [ "${SYSLOG}" ]
then
if [ "${LOGONLYERRORS}" ]
then
LOGLEVEL="e"
fi
fi
# check that MAX_JOBS is natural number > 0
# empty string means run all in parallel
if ! echo "${MAX_JOBS}" | grep -q -E '^[1-9][0-9]*$|^$'
then
_exit_err "Invalid max jobs value \"${MAX_JOBS}\""
fi
#
# Setup interval
#
if [ $# -ge 1 ]; then
export INTERVAL="$1"
shift
else
usage
fi
#
# Check for configuraton directory
#
[ -d "${CCOLLECT_CONF}" ] || _exit_err "No configuration found in " \
"\"${CCOLLECT_CONF}\" (is \$CCOLLECT_CONF properly set?)"
#
# Create (portable!) source "array"
#
export no_sources=0
if [ "${USE_ALL}" = 1 ]; then
#
# Get sources from source configuration
#
( cd "${CSOURCES}" && ls -1 > "${TMP}" ) || \
_exit_err "Listing of sources failed. Aborting."
while read -r tmp; do
eval export "source_${no_sources}=\"${tmp}\""
no_sources=$((no_sources + 1))
done < "${TMP}"
else
#
# Get sources from command line
#
while [ "$#" -ge 1 ]; do
eval "arg=\"\$1\""
shift
# arg is assigned in the eval above
# shellcheck disable=SC2154
eval export "source_${no_sources}=\"${arg}\""
no_sources="$((no_sources + 1))"
done
fi
#
# Need at least ONE source to backup
#
if [ "${no_sources}" -lt 1 ]; then
usage
else
_techo "${HALF_VERSION}: Beginning backup using interval ${INTERVAL}"
fi
#
# Look for pre-exec command (general)
#
if [ -x "${CPREEXEC}" ]; then
_techo "Executing ${CPREEXEC} ..."
"${CPREEXEC}"; ret=$?
_techo "Finished ${CPREEXEC} (return code: ${ret})."
[ "${ret}" -eq 0 ] || _exit_err "${CPREEXEC} failed. Aborting"
fi
################################################################################
#
# Let's do the backup - here begins the real stuff
#
# in PARALLEL mode:
# * create control pipe
# * determine number of jobs to start at once
if [ "${PARALLEL}" ]; then
mkfifo "${CONTROL_PIPE}"
# fd 5 is tied to control pipe
eval "exec 5<>'${CONTROL_PIPE}'"
TRAPFUNC="${TRAPFUNC}; rm -f \"${CONTROL_PIPE}\""
# shellcheck disable=SC2064
trap "${TRAPFUNC}" 0 1 2 15
# determine how much parallel jobs to prestart
if [ "${MAX_JOBS}" ]
then
if [ "${MAX_JOBS}" -le "${no_sources}" ]
then
prestart="${MAX_JOBS}"
else
prestart="${no_sources}"
fi
else
prestart=0
fi
fi
source_no=0
while [ "${source_no}" -lt "${no_sources}" ]; do
#
# Get current source
#
eval export name=\"\$source_${source_no}\"
source_no=$((source_no + 1))
#
# Start ourself, if we want parallel execution
#
if [ "${PARALLEL}" ]; then
if [ ! "${MAX_JOBS}" ]
then
# run all in parallel
"$0" "${INTERVAL}" "${name}" &
continue
elif [ "${prestart}" -gt 0 ]
then
# run prestart child if pending
{ "$0" "${INTERVAL}" "${name}"; printf '\n' >&5; } &
prestart=$((prestart - 1))
continue
else
# each time a child finishes we get a line from the pipe
# and then launch another child
while read -r line
do
{ "$0" "${INTERVAL}" "${name}"; printf '\n' >&5; } &
# get out of loop so we can contnue with main loop
# for next source
break
done <&5
continue
fi
fi
#
# Start subshell for easy log editing
#
(
backup="${CSOURCES}/${name}"
c_source="${backup}/source"
c_dest="${backup}/destination"
c_pre_exec="${backup}/pre_exec"
c_post_exec="${backup}/post_exec"
#
# Stderr to stdout, so we can produce nice logs
#
exec 2>&1
#
# Record start of backup: internal and for the user
#
begin_s="$(${SDATE})"
_techo "Beginning to backup"
#
# Standard configuration checks
#
if [ ! -e "${backup}" ]; then
_exit_err "Source \"${backup}\" does not exist."
fi
#
# Configuration _must_ be a directory (cconfig style)
#
if [ ! -d "${backup}" ]; then
_exit_err "\"${backup}\" is not a cconfig-directory. Skipping."
fi
#
# Acquire lock for source. If lock cannot be acquired, lock will exit
# with error message.
#
lock "${name}"
# redefine trap to also unlock (rm lockfile)
TRAPFUNC="${TRAPFUNC}; unlock \"${name}\""
# shellcheck disable=SC2064
trap "${TRAPFUNC}" 1 2 15
#
# First execute pre_exec, which may generate destination or other parameters
#
if [ -x "${c_pre_exec}" ]; then
_techo "Executing ${c_pre_exec} ..."
"${c_pre_exec}"; ret="$?"
_techo "Finished ${c_pre_exec} (return code ${ret})."
[ "${ret}" -eq 0 ] || _exit_err "${c_pre_exec} failed. Skipping."
fi
#
# Read source configuration
#
for opt in verbose very_verbose summary exclude rsync_options \
delete_incomplete rsync_failure_codes \
mtime quiet_if_down ; do
if [ -f "${backup}/${opt}" ] || [ -f "${backup}/no_${opt}" ]; then
eval "c_$opt=\"${backup}/$opt\""
else
eval "c_$opt=\"${CDEFAULTS}/$opt\""
fi
done
#
# Interval definition: First try source specific, fallback to default
#
c_interval="$(cat "${backup}/intervals/${INTERVAL}" 2>/dev/null)"
if [ -z "${c_interval}" ]; then
c_interval="$(cat "${CDEFAULTS}/intervals/${INTERVAL}" 2>/dev/null)"
if [ -z "${c_interval}" ]; then
_exit_err "No definition for interval \"${INTERVAL}\" found. Skipping."
fi
fi
#
# Sort by ctime (default) or mtime (configuration option)
#
# variable is assigned using eval
# shellcheck disable=SC2154
if [ -f "${c_mtime}" ] ; then
TSORT="t"
else
TSORT="tc"
fi
#
# Source configuration checks
#
if [ ! -f "${c_source}" ]; then
_exit_err "Source description \"${c_source}\" is not a file. Skipping."
else
source=$(cat "${c_source}"); ret="$?"
if [ "${ret}" -ne 0 ]; then
_exit_err "Source ${c_source} is not readable. Skipping."
fi
fi
#
# Destination is a path
#
if [ ! -f "${c_dest}" ]; then
_exit_err "Destination ${c_dest} is not a file. Skipping."
else
ddir="$(cat "${c_dest}")"; ret="$?"
if [ "${ret}" -ne 0 ]; then
_exit_err "Destination ${c_dest} is not readable. Skipping."
fi
fi
#
# Parameters: ccollect defaults, configuration options, user options
#
#
# Rsync standard options (archive will be added after is-up-check)
#
set -- "$@" "--delete" "--numeric-ids" "--relative" \
"--delete-excluded" "--sparse"
#
# Exclude list
#
# variable is assigned using eval
# shellcheck disable=SC2154
if [ -f "${c_exclude}" ]; then
set -- "$@" "--exclude-from=${c_exclude}"
fi
#
# Output a summary
#
# variable is assigned using eval
# shellcheck disable=SC2154
if [ -f "${c_summary}" ]; then
set -- "$@" "--stats"
fi
#
# Verbosity for rsync, rm, and mkdir
#
VVERBOSE=""
# variable is assigned using eval
# shellcheck disable=SC2154
if [ -f "${c_very_verbose}" ]; then
set -- "$@" "-vv"
VVERBOSE="-v"
elif [ -f "${c_verbose}" ]; then
set -- "$@" "-v"
fi
#
# Extra options for rsync provided by the user
#
# variable is assigned using eval
# shellcheck disable=SC2154
if [ -f "${c_rsync_options}" ]; then
while read -r line; do
# Trim line.
ln=$(echo "${line}" | awk '{$1=$1;print;}')
# Only if ln is non zero length string.
#
# If ln is empty then rsync '' DEST evaluates
# to transfer current directory to DEST which would
# with specific options destroy DEST content.
if [ -n "${ln}" ]
then
set -- "$@" "${ln}"
fi
done < "${c_rsync_options}"
fi
#
# Check: source is up and accepting connections (before deleting old backups!)
#
if ! rsync "$@" "${source}" >/dev/null 2>"${TMP}" ; then
# variable is assigned using eval
# shellcheck disable=SC2154
if [ ! -f "${c_quiet_if_down}" ]; then
cat "${TMP}"
fi
_exit_err "Source ${source} is not readable. Skipping."
fi
#
# Add --archive for real backup (looks nice in front)
#
set -- "--archive" "$@"
#
# Check: destination exists?
#
cd "${ddir}" || _exit_err "Cannot change to ${ddir}. Skipping."
#
# Check incomplete backups (needs echo to remove newlines)
#
# shellcheck disable=SC2010
ls -1 | grep "${CMARKER}\$" > "${TMP}"; ret=$?
if [ "$ret" -eq 0 ]; then
_techo "Incomplete backups: $(cat "${TMP}")"
# variable is assigned using eval
# shellcheck disable=SC2154
if [ -f "${c_delete_incomplete}" ]; then
delete_from_file "${TMP}" "${CMARKER}" &
fi
fi
#
# Include current time in name, not the time when we began to remove above
#
destination_name="${INTERVAL}.$(${CDATE}).$$-${source_no}"
export destination_name
destination_dir="${ddir}/${destination_name}"
export destination_dir
#
# Check: maximum number of backups is reached?
#
# shellcheck disable=SC2010
count="$(ls -1 | grep -c "^${INTERVAL}\\.")"
_techo "Existing backups: ${count} Total keeping backups: ${c_interval}"
if [ "${count}" -ge "${c_interval}" ]; then
# Use oldest directory as new backup destination directory.
# It need not to be deleted, rsync will sync its content.
# shellcheck disable=SC2010
oldest_bak=$(ls -${TSORT}1r | grep "^${INTERVAL}\\." | head -n 1 || \
_exit_err "Listing oldest backup failed")
_techo "Using ${oldest_bak} for destination dir ${destination_dir}"
if mv "${oldest_bak}" "${destination_dir}"; then
# Touch dest dir so it is not sorted wrong in listings below.
ls_rm_exclude=$(basename "${destination_dir}")
# We have something to remove only if count > interval.
remove="$((count - c_interval))"
else
_techo_err "Renaming oldest backup ${oldest_bak} to ${destination_dir} failed, removing it."
remove="$((count - c_interval + 1))"
ls_rm_exclude=""
fi
if [ "${remove}" -gt 0 ]; then
_techo "Removing ${remove} backup(s)..."
if [ -z "${ls_rm_exclude}" ]; then
# shellcheck disable=SC2010
ls -${TSORT}1r | grep "^${INTERVAL}\\." | head -n "${remove}" > "${TMP}" || \
_exit_err "Listing old backups failed"
else
# shellcheck disable=SC2010
ls -${TSORT}1r | grep -v "${ls_rm_exclude}" | grep "^${INTERVAL}\\." | head -n "${remove}" > "${TMP}" || \
_exit_err "Listing old backups failed"
fi
delete_from_file "${TMP}" &
fi
fi
#
# Check for backup directory to clone from: Always clone from the latest one!
# Exclude destination_dir from listing, it can be touched reused and renamed
# oldest existing destination directory.
#
dest_dir_name=$(basename "${destination_dir}")
# shellcheck disable=SC2010
last_dir="$(ls -${TSORT}p1 | grep '/$' | grep -v "${dest_dir_name}" | head -n 1)" || \
_exit_err "Failed to list contents of ${ddir}."
#
# Clone from old backup, if existing
#
if [ "${last_dir}" ]; then
set -- "$@" "--link-dest=${ddir}/${last_dir}"
_techo "Hard linking from ${last_dir}"
fi
#
# Mark backup running and go back to original directory
#
touch "${destination_dir}${CMARKER}"
cd "${__abs_mydir}" || _exit_err "Cannot go back to ${__abs_mydir}."
#
# the rsync part
#
_techo "Transferring files..."
rsync "$@" "${source}" "${destination_dir}"; ret=$?
_techo "Finished backup (rsync return code: $ret)."
#
# export rsync return code, might be useful in post_exec
#
export rsync_return_code=$ret
#
# Set modification time (mtime) to current time, if sorting by mtime is enabled
#
[ -f "$c_mtime" ] && touch "${destination_dir}"
#
# Check if rsync exit code indicates failure.
#
fail=""
# variable is assigned using eval
# shellcheck disable=SC2154
if [ -f "$c_rsync_failure_codes" ]; then
while read -r code ; do
if [ "$ret" = "$code" ]; then
fail=1
fi
done <"${c_rsync_failure_codes}"
fi
#
# Remove marking here unless rsync failed.
#
if [ -z "$fail" ]; then
rm "${destination_dir}${CMARKER}" || \
_exit_err "Removing ${destination_dir}${CMARKER} failed."
if [ "${ret}" -ne 0 ]; then
_techo "Warning: rsync exited non-zero, the backup may be broken (see rsync errors)."
fi
else
_techo "Warning: rsync failed with return code $ret."
fi
#
# Create symlink to newest backup
#
# shellcheck disable=SC2010
latest_dir="$(ls -${TSORT}p1 "${ddir}" | grep '/$' | head -n 1)" || \
_exit_err "Failed to list content of ${ddir}."
ln -snf "${ddir}${latest_dir}" "${ddir}current" || \
_exit_err "Failed to create 'current' symlink."
#
# post_exec
#
if [ -x "${c_post_exec}" ]; then
_techo "Executing ${c_post_exec} ..."
"${c_post_exec}"; ret=$?
_techo "Finished ${c_post_exec}."
if [ "${ret}" -ne 0 ]; then
_exit_err "${c_post_exec} failed."
fi
fi
#
# Time calculation
#
end_s="$(${SDATE})"
full_seconds="$((end_s - begin_s))"
hours="$((full_seconds / 3600))"
minutes="$(((full_seconds % 3600) / 60))"
seconds="$((full_seconds % 60))"
_techo "Backup lasted: ${hours}:${minutes}:${seconds} (h:m:s)"
unlock "${name}"
# wait for children (doing delete_from_file) if any still running
wait
) || exit
done
#
# Be a good parent and wait for our children, if they are running wild parallel
# After all children are finished then remove control pipe.
#
if [ "${PARALLEL}" ]; then
_techo "Waiting for children to complete..."
wait
rm -f "${CONTROL_PIPE}"
fi
#
# Look for post-exec command (general)
#
if [ -x "${CPOSTEXEC}" ]; then
_techo "Executing ${CPOSTEXEC} ..."
"${CPOSTEXEC}"; ret=$?
_techo "Finished ${CPOSTEXEC} (return code: ${ret})."
if [ "${ret}" -ne 0 ]; then
_techo "${CPOSTEXEC} failed."
fi
fi
rm -f "${TMP}"
_techo "Finished"

596
ccollect.sh Executable file
View file

@ -0,0 +1,596 @@
#!/bin/sh
#
# 2005-2009 Nico Schottelius (nico-ccollect at schottelius.org)
#
# This file is part of ccollect.
#
# ccollect is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# ccollect is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with ccollect. If not, see <http://www.gnu.org/licenses/>.
#
# Initially written for SyGroup (www.sygroup.ch)
# Date: Mon Nov 14 11:45:11 CET 2005
# Error upon expanding unset variables:
set -u
#
# Standard variables (stolen from cconf)
#
__pwd="$(pwd -P)"
__mydir="${0%/*}"; __abs_mydir="$(cd "$__mydir" && pwd -P)"
__myname=${0##*/}; __abs_myname="$__abs_mydir/$__myname"
#
# where to find our configuration and temporary file
#
CCOLLECT_CONF="${CCOLLECT_CONF:-/etc/ccollect}"
CSOURCES="${CCOLLECT_CONF}/sources"
CDEFAULTS="${CCOLLECT_CONF}/defaults"
CPREEXEC="${CDEFAULTS}/pre_exec"
CPOSTEXEC="${CDEFAULTS}/post_exec"
export TMP=$(mktemp "/tmp/${__myname}.XXXXXX")
VERSION="0.8"
RELEASE="2009-08-20"
HALF_VERSION="ccollect ${VERSION}"
FULL_VERSION="ccollect ${VERSION} (${RELEASE})"
#
# CDATE: how we use it for naming of the archives
# DDATE: how the user should see it in our output (DISPLAY)
#
CDATE="date +%Y%m%d-%H%M"
DDATE="date +%Y-%m-%d-%H:%M:%S"
SDATE="date +%s"
#
# unset values
#
PARALLEL=""
USE_ALL=""
#
# catch signals
#
trap "rm -f \"${TMP}\"" 1 2 15
#
# Functions
#
# time displaying echo
_techo()
{
echo "$(${DDATE}): $@"
}
# exit on error
_exit_err()
{
_techo "$@"
rm -f "${TMP}"
exit 1
}
add_name()
{
awk "{ print \"[${name}] \" \$0 }"
}
#
# Execute on remote host, if backing up to a remote host
#
pcmd()
{
if [ "${remote_host}" ]; then
ssh "${remote_host}" "$@"
else
"$@"
fi
}
delete_from_file()
{
#
# ssh-"feature": we cannot do '... read ...; ssh ...; < file',
# because ssh reads stdin! -n does not work -> does not ask for password
#
file="$1"; shift
while read to_remove; do set -- "$@" "${ddir}/${to_remove}"; done < "${file}"
_techo "Removing $@ ..."
pcmd rm ${VVERBOSE} -rf "$@" || _exit_err "Removing $@ failed."
}
display_version()
{
echo "${FULL_VERSION}"
exit 0
}
usage()
{
cat << eof
${__myname}: [args] <interval name> <sources to backup>
ccollect creates (pseudo) incremental backups
-h, --help: Show this help screen
-p, --parallel: Parallelise backup processes
-a, --all: Backup all sources specified in ${CSOURCES}
-v, --verbose: Be very verbose (uses set -x)
-V, --version: Print version information
This is version ${VERSION}, released on ${RELEASE}
(the first version was written on 2005-12-05 by Nico Schottelius).
Retrieve latest ccollect at http://www.nico.schottelius.org/software/ccollect/
eof
exit 0
}
#
# Parse options
#
while [ "$#" -ge 1 ]; do
case "$1" in
-a|--all)
USE_ALL=1
;;
-v|--verbose)
set -x
;;
-p|--parallel)
PARALLEL=1
;;
-h|--help)
usage
;;
-V|--version)
display_version
;;
-h|--help|-*)
usage
;;
--)
# ignore the -- itself
shift
break
;;
*)
break
;;
esac
shift
done
#
# Setup interval
#
if [ $# -ge 1 ]; then
export INTERVAL="$1"
shift
else
usage
fi
#
# Check for configuraton directory
#
[ -d "${CCOLLECT_CONF}" ] || _exit_err "No configuration found in " \
"\"${CCOLLECT_CONF}\" (is \$CCOLLECT_CONF properly set?)"
#
# Create (portable!) source "array"
#
export no_sources=0
if [ "${USE_ALL}" = 1 ]; then
#
# Get sources from source configuration
#
( cd "${CSOURCES}" && ls -1 > "${TMP}" ); ret=$?
[ "${ret}" -eq 0 ] || _exit_err "Listing of sources failed. Aborting."
while read tmp; do
eval export source_${no_sources}=\"${tmp}\"
no_sources=$((${no_sources}+1))
done < "${TMP}"
else
#
# Get sources from command line
#
while [ "$#" -ge 1 ]; do
eval arg=\"\$1\"; shift
eval export source_${no_sources}=\"${arg}\"
no_sources="$((${no_sources}+1))"
done
fi
#
# Need at least ONE source to backup
#
if [ "${no_sources}" -lt 1 ]; then
usage
else
_techo "${HALF_VERSION}: Beginning backup using interval ${INTERVAL}"
fi
#
# Look for pre-exec command (general)
#
if [ -x "${CPREEXEC}" ]; then
_techo "Executing ${CPREEXEC} ..."
"${CPREEXEC}"; ret=$?
_techo "Finished ${CPREEXEC} (return code: ${ret})."
[ "${ret}" -eq 0 ] || _exit_err "${CPREEXEC} failed. Aborting"
fi
#
# Let's do the backup
#
i=0
while [ "${i}" -lt "${no_sources}" ]; do
#
# Get current source
#
eval name=\"\$source_${i}\"
i=$((${i}+1))
export name
#
# start ourself, if we want parallel execution
#
if [ "${PARALLEL}" ]; then
"$0" "${INTERVAL}" "${name}" &
continue
fi
#
# Start subshell for easy log editing
#
(
backup="${CSOURCES}/${name}"
#
# Stderr to stdout, so we can produce nice logs
#
exec 2>&1
#
# Record start of backup: internal and for the user
#
begin_s="$(${SDATE})"
_techo "Beginning to backup"
#
# Standard configuration checks
#
if [ ! -e "${backup}" ]; then
_exit_err "Source does not exist."
fi
#
# Configuration _must_ be a directory (cconfig style)
#
if [ ! -d "${backup}" ]; then
_exit_err "\"${name}\" is not a cconfig-directory. Skipping."
fi
#
# Read / create configuration
#
c_source="${backup}/source"
c_dest="${backup}/destination"
c_pre_exec="${backup}/pre_exec"
c_post_exec="${backup}/post_exec"
c_marker="ccollect-marker"
for opt in verbose very_verbose summary exclude rsync_options \
delete_incomplete remote_host rsync_failure_codes \
mtime quiet_if_down ; do
if [ -f "${backup}/${opt}" -o -f "${backup}/no_${opt}" ]; then
eval c_$opt=\"${backup}/$opt\"
else
eval c_$opt=\"${CDEFAULTS}/$opt\"
fi
done
#
# Sort by ctime (default) or mtime (configuration option)
#
if [ -f "$c_mtime" ] ; then
TSORT="t"
else
TSORT="tc"
fi
#
# First execute pre_exec, which may generate destination or other parameters
#
if [ -x "${c_pre_exec}" ]; then
_techo "Executing ${c_pre_exec} ..."
"${c_pre_exec}"; ret="$?"
_techo "Finished ${c_pre_exec} (return code ${ret})."
if [ "${ret}" -ne 0 ]; then
_exit_err "${c_pre_exec} failed. Skipping."
fi
fi
#
# Source configuration checks
#
if [ ! -f "${c_source}" ]; then
_exit_err "Source description \"${c_source}\" is not a file. Skipping."
else
source=$(cat "${c_source}"); ret="$?"
if [ "${ret}" -ne 0 ]; then
_exit_err "Source ${c_source} is not readable. Skipping."
fi
fi
#
# Destination is a path
#
if [ ! -f "${c_dest}" ]; then
_exit_err "Destination ${c_dest} is not a file. Skipping."
else
ddir="$(cat "${c_dest}")"; ret="$?"
if [ "${ret}" -ne 0 ]; then
_exit_err "Destination ${c_dest} is not readable. Skipping."
fi
fi
#
# Set pre-cmd, if we backup to a remote host.
#
if [ -f "${c_remote_host}" ]; then
remote_host="$(cat "${c_remote_host}")"; ret="$?"
if [ "${ret}" -ne 0 ]; then
_exit_err "Remote host file ${c_remote_host} exists, but is not readable. Skipping."
fi
destination="${remote_host}:${ddir}"
else
remote_host=""
destination="${ddir}"
fi
export remote_host
#
# Parameters: ccollect defaults, configuration options, user options
#
#
# Rsync standard options
#
set -- "$@" "--archive" "--delete" "--numeric-ids" "--relative" \
"--delete-excluded" "--sparse"
#
# Exclude list
#
if [ -f "${c_exclude}" ]; then
set -- "$@" "--exclude-from=${c_exclude}"
fi
#
# Output a summary
#
if [ -f "${c_summary}" ]; then
set -- "$@" "--stats"
fi
#
# Verbosity for rsync, rm, and mkdir
#
VVERBOSE=""
if [ -f "${c_very_verbose}" ]; then
set -- "$@" "-vv"
VVERBOSE="-v"
elif [ -f "${c_verbose}" ]; then
set -- "$@" "-v"
fi
#
# Extra options for rsync provided by the user
#
if [ -f "${c_rsync_options}" ]; then
while read line; do
set -- "$@" "$line"
done < "${c_rsync_options}"
fi
#
# Check: source is up and accepting connections (before deleting old backups!)
#
if ! rsync "${source}" >/dev/null 2>"${TMP}" ; then
if [ ! -f "${c_quiet_if_down}" ]; then
cat "${TMP}"
fi
_exit_err "Source ${source} is not readable. Skipping."
fi
#
# Check: destination exists?
#
( pcmd cd "${ddir}" ) || _exit_err "Cannot change to ${ddir}. Skipping."
#
# Check: incomplete backups? (needs echo to remove newlines)
#
incomplete="$(echo \
$(pcmd ls -1 "${ddir}/" | \
awk "/\.${c_marker}\$/ { print \$0; gsub(\"\.${c_marker}\$\",\"\",\$0); print \$0 }" | \
tee "${TMP}"))"
if [ "${incomplete}" ]; then
_techo "Incomplete backups: ${incomplete}"
if [ -f "${c_delete_incomplete}" ]; then
delete_from_file "${TMP}"
fi
fi
#
# Interval definition: First try source specific, fallback to default
#
c_interval="$(cat "${backup}/intervals/${INTERVAL}" 2>/dev/null)"
if [ -z "${c_interval}" ]; then
c_interval="$(cat "${CDEFAULTS}/intervals/${INTERVAL}" 2>/dev/null)"
if [ -z "${c_interval}" ]; then
_exit_err "No definition for interval \"${INTERVAL}\" found. Skipping."
fi
fi
#
# Check: maximum number of backups is reached?
# If so remove. Use grep and ls -p so we only look at directories
#
count="$(pcmd ls -p1 "${ddir}" | grep "^${INTERVAL}\..*/\$" | wc -l \
| sed 's/^ *//g')" || _exit_err "Counting backups failed"
_techo "Existing backups: ${count} Total keeping backups: ${c_interval}"
if [ "${count}" -ge "${c_interval}" ]; then
substract="$((${c_interval} - 1))"
remove="$((${count} - ${substract}))"
_techo "Removing ${remove} backup(s)..."
pcmd ls -${TSORT}p1r "${ddir}" | grep "^${INTERVAL}\..*/\$" | \
head -n "${remove}" > "${TMP}" || \
_exit_err "Listing old backups failed"
delete_from_file "${TMP}"
fi
#
# Check for backup directory to clone from: Always clone from the latest one!
#
last_dir="$(pcmd ls -${TSORT}p1 "${ddir}" | grep '/$' | head -n 1)" || \
_exit_err "Failed to list contents of ${ddir}."
#
# clone from old backup, if existing
#
if [ "${last_dir}" ]; then
set -- "$@" "--link-dest=${ddir}/${last_dir}"
_techo "Hard linking from ${last_dir}"
fi
# set time when we really begin to backup, not when we began to remove above
destination_date="$(${CDATE})"
destination_dir="${ddir}/${INTERVAL}.${destination_date}.$$"
destination_full="${destination}/${INTERVAL}.${destination_date}.$$"
# give some info
_techo "Beginning to backup, this may take some time..."
_techo "Creating ${destination_dir} ..."
pcmd mkdir ${VVERBOSE} "${destination_dir}" || \
_exit_err "Creating ${destination_dir} failed. Skipping."
#
# added marking in 0.6 (and remove it, if successful later)
#
pcmd touch "${destination_dir}.${c_marker}"
#
# the rsync part
#
_techo "Transferring files..."
rsync "$@" "${source}" "${destination_full}"; ret=$?
_techo "Finished backup (rsync return code: $ret)."
#
# Set modification time (mtime) to current time, if sorting by mtime is enabled
#
[ -f "$c_mtime" ] && pcmd touch "${destination_dir}"
#
# Check if rsync exit code indicates failure.
#
fail=""
if [ -f "$c_rsync_failure_codes" ]; then
while read code ; do
if [ "$ret" = "$code" ]; then
fail=1
fi
done <"$c_rsync_failure_codes"
fi
#
# Remove marking here unless rsync failed.
#
if [ -z "$fail" ]; then
pcmd rm "${destination_dir}.${c_marker}" || \
_exit_err "Removing ${destination_dir}.${c_marker} failed."
if [ "${ret}" -ne 0 ]; then
_techo "Warning: rsync exited non-zero, the backup may be broken (see rsync errors)."
fi
else
_techo "Warning: rsync failed with return code $ret."
fi
#
# post_exec
#
if [ -x "${c_post_exec}" ]; then
_techo "Executing ${c_post_exec} ..."
"${c_post_exec}"; ret=$?
_techo "Finished ${c_post_exec}."
if [ "${ret}" -ne 0 ]; then
_exit_err "${c_post_exec} failed."
fi
fi
# Calculation
end_s="$(${SDATE})"
full_seconds="$((${end_s} - ${begin_s}))"
hours="$((${full_seconds} / 3600))"
seconds="$((${full_seconds} - (${hours} * 3600)))"
minutes="$((${seconds} / 60))"
seconds="$((${seconds} - (${minutes} * 60)))"
_techo "Backup lasted: ${hours}:${minutes}:${seconds} (h:m:s)"
) | add_name
done
#
# Be a good parent and wait for our children, if they are running wild parallel
#
if [ "${PARALLEL}" ]; then
_techo "Waiting for children to complete..."
wait
fi
#
# Look for post-exec command (general)
#
if [ -x "${CPOSTEXEC}" ]; then
_techo "Executing ${CPOSTEXEC} ..."
"${CPOSTEXEC}"; ret=$?
_techo "Finished ${CPOSTEXEC} (return code: ${ret})."
if [ "${ret}" -ne 0 ]; then
_techo "${CPOSTEXEC} failed."
fi
fi
rm -f "${TMP}"
_techo "Finished"

View file

@ -0,0 +1 @@
/tmp/ccollect

View file

@ -0,0 +1 @@
localhost

View file

@ -0,0 +1 @@
/home/users/nico/bin

View file

@ -1,79 +0,0 @@
Summary: (pseudo) incremental backup with different exclude lists using hardlinks and rsync
Name: ccollect
Version: 2.3
Release: 0
URL: http://www.nico.schottelius.org/software/ccollect
Source0: http://www.nico.schottelius.org/software/ccollect/%{name}-%{version}.tar.bz2
License: GPL-3
Group: Applications/System
Vendor: Nico Schottelius <nico-ccollect@schottelius.org>
BuildRoot: %{_tmppath}/%{name}-%(id -un)
BuildArch: noarch
Requires: rsync
%description
Ccollect backups data from local and remote hosts to your local harddisk.
Although ccollect creates full backups, it requires very less space on the backup medium, because ccollect uses hardlinks to create an initial copy of the last backup.
Only the inodes used by the hardlinks and the changed files need additional space.
%prep
%setup -q
%install
rm -rf $RPM_BUILD_ROOT
#Installing main ccollect and /etc directory
%__install -d 755 %buildroot%_bindir
%__install -d 755 %buildroot%_sysconfdir/%name
%__install -m 755 ccollect %buildroot%_bindir/
#bin files from tools directory
for t in $(ls tools/ccollect_*) ; do
%__install -m 755 ${t} %buildroot%_bindir/
done
#Configuration examples and docs
%__install -d 755 %buildroot%_datadir/doc/%name-%version/examples
%__install -m 644 README %buildroot%_datadir/doc/%name-%version
%__install -m 644 COPYING %buildroot%_datadir/doc/%name-%version
%__install -m 644 CREDITS %buildroot%_datadir/doc/%name-%version
%__install -m 644 conf/README %buildroot%_datadir/doc/%name-%version/examples
%__cp -pr conf/defaults %buildroot%_datadir/doc/%name-%version/examples/
%__cp -pr conf/sources %buildroot%_datadir/doc/%name-%version/examples/
#Addition documentation and some config tools
%__install -d 755 %buildroot%_datadir/%name/tools
%__install -m 755 tools/called_from_remote_pre_exec %buildroot%_datadir/%name/tools
%__cp -pr tools/config-pre-* %buildroot%_datadir/%name/tools
%__install -m 755 tools/report_success %buildroot%_datadir/%name/tools
%clean
rm -rf $RPM_BUILD_ROOT
%files
%defattr(-,root,root)
%_bindir/ccollect*
%_datadir/doc/%name-%version
%_datadir/%name/tools
%docdir %_datadir/doc/%name-%version
%dir %_sysconfdir/%name
%changelog
* Thu Aug 20 2009 Nico Schottelius <nico-ccollect@schottelius.org> 0.8
- Introduce consistenst time sorting (John Lawless)
- Check for source connectivity before trying backup (John Lawless)
- Defensive programming patch (John Lawless)
- Some code cleanups (argument parsing, usage) (Nico Schottelius)
- Only consider directories as sources when using -a (Nico Schottelius)
- Fix general parsing problem with -a (Nico Schottelius)
- Fix potential bug when using remote_host, delete_incomplete and ssh (Nico Schottelius)
- Improve removal performance: minimised number of 'rm' calls (Nico Schottelius)
- Support sorting by mtime (John Lawless)
- Improve option handling (John Lawless)
- Add support for quiet operation for dead devices (quiet_if_down) (John Lawless)
- Add smart option parsing, including support for default values (John Lawless)
- Updated and cleaned up documentation (Nico Schottelius)
- Fixed bug "removal of current directory" in ccollect_delete_source.sh (Found by G????nter St????hr, fixed by Nico Schottelius)

View file

@ -1,47 +0,0 @@
[Almost complete Copy of an e-mail from Patrick Drolet]
Hello again,
I have created a script to better manage the backups since my
upload/download ratio and my bandwidth is limited by my ISP, and my hard
disk space is also somewhat limited. The script is called
"ccollect_mgr.sh".
Provides the following features
1) Determine the interval (daily/weekly/monthly)
a. Define when you want weekly and monthly backups. It takes care of
the rest
2) Perform the backups using ccollect
3) Copy the ccollect log output to the first backup of the set
a. Keeping the detailed log of each backup is always handy!
4) Build a periodic report and include the real amount of disk used
a. Computes the real amount of disk used (eg: no double counting of
hard links)
b. Shows the actual amount of data transferred
5) Send an email if there has been errors or warnings
6) Send a periodic email to show transfer size, real backup size, etc
a. Weekly reports are nice.!
[...]
- rdu (real du), which computes the real amount of disk used (no
double/triple counting hard links), same code as in ccollect_mgr.sh.
- S60ccollect_example, an example script to put in etc/init.d to
add ccollect_mgr to the crontab

View file

@ -1,21 +0,0 @@
#!/bin/sh
# Standard Linux: put in /etc/init.d
# Busybox: put in /opt/etc/init.d
# Add ccollect_mgr job to crontab
# Syntax reminder from crontab:
# minute 0-59
# hour 0-23
# day of month 1-31
# month 1-12 (or names, see below)
# day of week 0-7 (0 or 7 is Sun, or use names)
crontab -l | grep -v ccollect_mgr > /tmp/crontab.tmp
# Backup every day at 1 am.
echo "00 01 * * * /usr/local/sbin/ccollect_mgr.sh -from nas@myemail.net -to me@myemail.net -server relay_or_smtp_server NAS > /usr/local/var/log/ccollect.cron &" >> /tmp/crontab.tmp
crontab /tmp/crontab.tmp
rm /tmp/crontab.tmp

View file

@ -1,542 +0,0 @@
#!/bin/sh
#
# ----------------------------------------------------------------------------
# Last update: 2009-12-11
# By : pdrolet (ccollect_mgr@drolet.name)
# ----------------------------------------------------------------------------
# Job manager to the ccollect utilities
# (ccollect written by Nico Schottelius)
#
# Provides the following features
# 1) Determine the interval (daily/weekly/monthly)
# 2) Check the estimated file transfer size
# 3) Perform the backups using ccollect
# 4) Copy the ccollect log to the first backup of the set
# 5) Build a periodic report and include the real amount of disk used
# 6) Send an email if there has been errors or warnings
# 7) Send a periodic email to show transfer size, real backup size, etc
# ----------------------------------------------------------------------------
#
# This script was written primarily to gain better visibility of backups in
# an environment where data transfer is limited and so is bandwidth
# (eg: going through an ISP). The primary target of this script were a
# DNS323 and a QNAP T209 (eg: Busybox devices and not standard Linux devices)
# but it should run on any POSIX compliant device.
#
# Note: This is one of my first script in over a decade... don't use this as a
# reference (but take a look at ccollect.sh... very well written!)
# ----------------------------------------------------------------------------
#
# -------------------------------------------
# TO MAKE THIS SCRIPT RUN ON A BUSYBOX DEVICE
# -------------------------------------------
# - You may need to install Optware and the following packages:
# - findutils (to get a find utility which supports printf)
# - procps (to get a ps utility that is standard)
# - mini-sendmail (this is what I used to send emails... you could easily
# modify this to use sendmail, mutt, putmail, etc...).
# - On DNS323 only: Your Busybox is very limited. For details, see
# http://wiki.dns323.info/howto:ffp#shells. You need to redirect /bin/sh
# to the Busybox provided with ffp (Fun Plug). To do this, type:
# ln -fs /ffp/bin/sh /bin/sh
#
# --------------------------------------------------
# TO MAKE THIS SCRIPT RUN ON A STANDARD LINUX DEVICE
# --------------------------------------------------
# - You will need install mini_sendmail or rewrite the send_email routine.
#
# ----------------------------------------------------------------------------
# Send warning if the worst case data transfer will be larger than (in MB)...
warning_transfer_size=1024
abort_transfer_size=5120
# Define paths and default file names
ADD_TO_PATH="/opt/bin:/opt/sbin:/usr/local/bin:/usr/local/sbin"
CCOLLECT="ccollect.sh"
CCOLLECT_CONF="/usr/local/etc/ccollect"
PS="/opt/bin/ps"
FIND="/opt/bin/find"
TEMP_LOG="${CCOLLECT_CONF}"/log.$$
per_report="${CCOLLECT_CONF}/periodic_report.log"
tmp_report="/tmp/ccollect.$$"
tmp_mgr="/tmp/ccollect_mgr.$$"
tmp_email="/tmp/email.$$"
backups_not_found=""
# Sub routines...
send_email()
{
# Send a simple email using mini-sendmail.
msg_body_file="$1"
shift
# ------------------------------
# Quit if we can't send an email
# ------------------------------
if [ "${to}" == "" ] || [ "${mail_server}" == "" ]; then
echo "Missing mail server or destination email. No email sent with subject: $@"
exit 1
fi
echo from: "${from}" > "${tmp_email}"
echo subject: "$@" >> "${tmp_email}"
echo to: "${to}" >> "${tmp_email}"
echo cc: >> "${tmp_email}"
echo bcc: >> "${tmp_email}"
echo "" >> "${tmp_email}"
echo "" >> "${tmp_email}"
cat "${msg_body_file}" >> "${tmp_email}"
echo "" >> "${tmp_email}"
echo ""
echo Sending email to ${to} to report the following:
echo -----------------------------------------------
cat "${tmp_email}"
cat "${tmp_email}" | mini_sendmail -f"${from}" -s"${mail_server}" "${to}"
rm "${tmp_email}"
}
remove_source()
{
remove_no=$1
eval echo Removing backup \"\$source_$1\"
no_sources="$(( ${no_sources} - 1 ))"
while [ "${remove_no}" -lt "${no_sources}" ]; do
eval source_${remove_no}=\"\$source_$(( ${remove_no} + 1))\"
eval ddir_${remove_no}=\"\$ddir_$(( ${remove_no} + 1))\"
remove_no=$(( ${remove_no} + 1 ))
done
}
compute_rdu()
{
kdivider=1
find_options=""
while [ "$#" -ge 1 ]; do
case "$1" in
-m)
kdivider=1024
;;
-g)
kdivider=1048576
;;
*)
break
;;
esac
shift
done
if [ "$#" == 0 ]; then
rdu=0
return 1
fi
# ------------------------------------------------------------------------------------------------------
# Compute the real disk usage (eg: hard links do files outside the backup set don't count)
# ------------------------------------------------------------------------------------------------------
# 1) Find selected files and list link count, inodes, file type and size
# 2) Sort (sorts on inodes since link count is constant per inode)
# 3) Merge duplicates using uniq
# (result is occurence count, link count, inode, file type and size)
# 4) Use awk to sum up the file size of each inodes when the occurence count
# and link count are the same. Use %k for size since awk's printf is 32 bits
# 5) Present the result with additional dividers based on command line parameters
#
rdu=$(( ( `"${FIND}" "$@" -printf '%n %i %y %k \n' \
| sort -n \
| uniq -c \
| awk '{ if (( $1 == $2 ) || ($4 == "d")) { sum += $5; } } END { printf "%u\n",(sum); }'` \
+ ${kdivider} - 1 ) / ${kdivider} ))
}
check_running_backups()
{
# Check if a backup is already ongoing. If so, skip and send email
# Don't use the ccollect marker as this is no indication if it is still running
source_no=0
while [ "${source_no}" -lt "${no_sources}" ]; do
eval backup=\"\$source_${source_no}\"
PID=$$
"${PS}" -e -o pid,ppid,args 2> /dev/null \
| grep -v -e grep -e "${PID}.*ccollect.*${backup}" \
| grep "ccollect.*${backup}" > "${tmp_mgr}" 2> /dev/null
running_proc=`cat "${tmp_mgr}" | wc -l`
if [ ${running_proc} -gt 0 ]; then
# Remove backup from list
running_backups="${running_backups}${backup} "
echo "Process already running:"
cat "${tmp_mgr}"
remove_source ${source_no}
else
source_no=$(( ${source_no} + 1 ))
fi
rm "${tmp_mgr}"
done
if [ "${running_backups}" != "" ]; then
echo "skipping ccollect backups already running: ${running_backups}" | tee "${tmp_report}"
send_email "${tmp_report}" "WARNING - skipping ccollect backups already running: ${running_backups}"
rm "${tmp_report}"
fi
}
find_interval()
{
# ----------------------------------------------------
# Find interval for ccollect backup.
# optional parameters:
# - Day of the week to do weekly backups
# - Do monthly instead of weekly on the Nth week
# ----------------------------------------------------
weekly_backup="$1"
monthly_backup="$2"
weekday=`date "+%w"`
if [ ${weekday} -eq ${weekly_backup} ]; then
dom=`date "+%e"`
weeknum=$(( ( ${dom} / 7 ) + 1 ))
if [ "${weeknum}" -eq "${monthly_backup}" ]; then
interval=monthly
else
interval=weekly
fi
else
interval=daily
fi
}
precheck_transfer_size()
{
# Check the estimated (worst case) transfer size and send email if larger than certain size
# Abort backup if total transfer is larger than maximum limit (ex: an error somewhere
# requires to do full backup and not incremental, which could blow the quota with ISP)
#
# Be nice and add error checking one day...
source_no=0
while [ "${source_no}" -lt "${no_sources}" ]; do
eval backup=\"\$source_${source_no}\"
eval ddir=\"\$ddir_${source_no}\"
last_dir="$(ls -tcp1 "${ddir}" | grep '/$' | head -n 1)"
sdir="$(cat "${CCOLLECT_CONF}"/sources/"${backup}"/source)"; ret="$?"
if [ -f "${CCOLLECT_CONF}"/sources/"${backup}"/exclude ]; then
exclude="--exclude-from=${CCOLLECT_CONF}/sources/${backup}/exclude";
else
exclude=""
fi
rsync_options=""
if [ -f "${CCOLLECT_CONF}"/sources/"${backup}"/rsync_options ]; then
while read line; do
rsync_options="${rsync_options} ${line}"
done < ${CCOLLECT_CONF}/sources/${backup}/rsync_options
fi
rsync -n -a --delete --stats ${rsync_options} "${exclude}" "${sdir}" "${ddir}/${last_dir}" > "${tmp_report}"
tx_rx=`cat "${tmp_report}" | grep "Total transferred file size" | \
awk '{ { tx += $5 } } END { printf "%u",(((tx)+1024*1024-1)/1024/1024); }'`
total_xfer=$(( ${total_xfer} + ${tx_rx} ))
source_no=$(( ${source_no} + 1 ))
done
echo "Transfer estimation for${ccollect_backups}: ${total_xfer} MB"
if [ ${total_xfer} -gt ${abort_transfer_size} ]; then
# --------------------------------------------------
# Send an error if transfer is larger than max limit
# --------------------------------------------------
# Useful to detect potential issues when there is transfer quota (ex: with ISP)
echo "Data transfer larger than ${abort_transfer_size} MB is expected for${ccollect_backups}" >> "${tmp_report}"
echo "** BACKUP ABORTED **" >> "${tmp_report}"
send_email "${tmp_report}" "ERROR: aborted ccollect for${ccollect_backups} -- Estimated Tx+Rx: ${total_xfer} MB"
rm "${tmp_report}"
exit 1
elif [ ${total_xfer} -gt ${warning_transfer_size} ]; then
# --------------------------------------------------
# Send a warning if transfer is expected to be large
# --------------------------------------------------
# Useful to detect potential issues when there is transfer quota (ex: with ISP)
echo "Data transfer larger than ${warning_transfer_size} MB is expected for${ccollect_backups}" > "${tmp_report}"
send_email "${tmp_report}" "WARNING ccollect for${ccollect_backups} -- Estimated Tx+Rx: ${total_xfer} MB"
rm "${tmp_report}"
fi
}
build_backup_dir_list()
{
source_no=0
while [ "${source_no}" -lt "${no_sources}" ]; do
eval backup=\"\$source_${source_no}\"
eval ddir=\"\$ddir_${source_no}\"
backup_dir="`cat "${TEMP_LOG}" \
| grep "\[${backup}\] .*: Creating.* ${ddir}" \
| head -n 1 \
| sed 's/[^\/]*\//\//; s/ \.\.\.//'`"
if [ ! -d "${backup_dir}" ]; then
backups_not_found="${backups_not_found}\"${backup}\" "
echo -n "Backup directory for \"${backup}\" not found. "
remove_source "${source_no}"
else
eval export backup_dir_list_${source_no}="${backup_dir}"
# eval echo Backup Dir List: \"\$backup_dir_list_${source_no}\"
source_no=$(( ${source_no} + 1 ))
fi
done
}
move_log()
{
if [ "${no_sources}" -gt 0 ]; then
eval log_file=\"\$backup_dir_list_1\"/ccollect.log
mv "${TEMP_LOG}" "${log_file}"
echo New Log Location: "${log_file}"
else
echo "WARNING: none of the backup set have been created"
log_file="${TEMP_LOG}"
fi
}
send_report()
{
# Analyze log for periodic report and for error status report
cat "${log_file}" | ccollect_analyse_logs.sh iwe > "${tmp_report}"
# -------------------------
# Build the periodic report
# -------------------------
# Compute the total number of MB sent and received for all the backup sets
tx_rx=`cat "${tmp_report}" | \
grep 'sent [[:digit:]]* bytes received [0-9]* bytes' | \
awk '{ { tx += $3 } { rx += $6} } END \
{ printf "%u",(((tx+rx)+(1024*1024)-1)/1024/1024); }'`
current_date=`date +'20%y/%m/%d %Hh%M -- '`
# ------------------------------------------
# Get the real disk usage for the backup set
# ------------------------------------------
total_rdu=0
source_no=0
while [ "${source_no}" -lt "${no_sources}" ]; do
eval backup_dir=\"\$backup_dir_list_${source_no}\"
compute_rdu -m "${backup_dir}"
total_rdu=$(( ${total_rdu} + ${rdu} ))
source_no=$(( ${source_no} + 1 ))
done
# ---------------------------------------------------------
# Get the disk usage for all backups of each backup sets...
# ** BE PATIENT!!! **
# ---------------------------------------------------------
historical_rdu=0
source_no=0
while [ "${source_no}" -lt "${no_sources}" ]; do
eval backup_dir=\"\$ddir_${source_no}\"
compute_rdu -m "${backup_dir}"
historical_rdu=$(( ${historical_rdu} + ${rdu} ))
source_no=$(( ${source_no} + 1 ))
done
historical_rdu=$(( (${historical_rdu}+1023) / 1024 ))
if [ "${no_sources}" -gt 0 ]; then
ccollect_backups=""
else
ccollect_backups="(none performed) "
fi
source_no=0
while [ "${source_no}" -lt "${no_sources}" ]; do
eval backup=\"\$source_${source_no}\"
ccollect_backups="${ccollect_backups}\"${backup}\" "
source_no=$(( ${source_no} + 1 ))
done
echo ${current_date} Tx+Rx: ${tx_rx} MB -- \
Disk Usage: ${total_rdu} MB -- \
Backup set \(${interval}\):${ccollect_backups} -- \
Historical backups usage: ${historical_rdu} GB >> "${per_report}"
echo "Total Data Transfer: ${tx_rx} MB -- Total Disk Usage: ${total_rdu} MB -- Total Historical backups usage: ${historical_rdu} GB"
# ----------------------------------------
# Send a status email if there is an error
# ----------------------------------------
ccollect_we=`cat "${log_file}" | ccollect_analyse_logs.sh we | wc -l`
if [ ${ccollect_we} -ge 1 ]; then
send_email "${tmp_report}" "ERROR ccollect for${ccollect_backups} -- Tx+Rx: ${tx_rx} MB"
fi
# --------------------
# Send periodic report
# --------------------
if [ ${report_interval} == ${interval} ] || [ ${interval} == "monthly" ]; then
# Make reporting atomic to handle concurrent ccollect_mgr instances
mv "${per_report}" "${per_report}".$$
cat "${per_report}".$$ >> "${per_report}".history
# Calculate total amount of bytes sent and received
tx_rx=`cat "${per_report}".$$ | \
awk '{ { transfer += $5 } } END \
{ printf "%u",(transfer); }'`
# Send email
send_email "${per_report}.$$" "${report_interval} ccollect status for${ccollect_backups} -- Tx+Rx: ${tx_rx} MB"
rm "${per_report}.$$"
fi
rm "${tmp_report}"
}
# ------------------------------------------------
# Add to PATH in case we're launching from crontab
# ------------------------------------------------
PATH="${ADD_TO_PATH}:${PATH}"
# --------------
# Default Values
# --------------
# Set on which interval status emails are sent (daily, weekly, monthly)
report_interval=weekly
# Set day of the week for weekly backups. Default is Monday
# 0=Sun, 1=Mon, 2=Tue, 3=Wed, 4=Thu, 5=Fri, 6=Sat
weekly_backup=1
# Set the monthly backup interval. Default is 4th Monday of every month
monthly_backup=4
# ---------------------------------
# Parse command line
# ---------------------------------
show_help=0
export no_sources=0
while [ "$#" -ge 1 ]; do
case "$1" in
-help)
show_help=1
;;
-from)
from="$2"
shift
;;
-to)
to="$2"
shift
;;
-server|mail_server)
mail_server="$2"
shift
;;
-weekly)
weekly_backup="$2"
shift
;;
-monthly)
monthly_backup="$2"
shift
;;
-warning_size)
warning_transfer_size="$2"
shift
;;
-abort_size)
abort_transfer_size="$2"
shift
;;
-report)
report_interval="$2"
shift
;;
-*)
ccollect_options="${ccollect_options}$1 "
;;
daily|weekly|monthly)
;;
*)
eval backup=\"\$1\"
ddir="$(cat "${CCOLLECT_CONF}"/sources/"${backup}"/destination)"; ret="$?"
if [ "${ret}" -ne 0 ]; then
echo "Destination ${CCOLLECT_CONF}/sources/${backup}/destination is not readable... Skipping."
else
ccollect_backups="${ccollect_backups} \"$1\""
eval export source_${no_sources}=\"\$1\"
eval export ddir_${no_sources}="${ddir}"
# eval echo Adding source \"\$source_${no_sources}\" -- \"\$ddir_${no_sources}\"
no_sources="$(( ${no_sources} + 1 ))"
fi
;;
esac
shift
done
if [ "${no_sources}" -lt 1 ] || [ ${show_help} -eq 1 ]; then
echo ""
echo "$0: Syntax"
echo " -help This help"
echo " -from <email> From email address (ex.: -from nas@home.com)"
echo " -to <email> Send email to this address (ex.: -to me@home.com)"
echo " -server <smtp_addr> SMTP server used for sending emails"
echo " -weekly <day#> Define wich day of the week is the weekly backup"
echo " Default is ${weekly_backup}. Sunday = 0, Saturday = 6"
echo " -monthly <week#> Define on which week # is the monthly backup"
echo " Default is ${monthly_backup}. Value = 1 to 5"
echo " -report <interval> Frequency of report email (daily, weekly or monthly)"
echo " Default is ${report_interval}"
echo " -warning_size <MB> Send a warning email if the transfer size exceed this"
echo " Default is ${warning_transfer_size} MB"
echo " -abort_size <MB> Abort and send an error email if the transfer size exceed this"
echo " Default is ${abort_transfer_size} MB"
echo ""
echo " other parameters are transfered to ccollect"
echo ""
exit 0
fi
#echo Backup sets:"${ccollect_backups}"
check_running_backups
if [ "${no_sources}" -lt 1 ]; then
echo "No backup sets are reachable"
exit 1
fi
find_interval ${weekly_backup} ${monthly_backup}
echo Interval: ${interval}
precheck_transfer_size
"${CCOLLECT}" ${ccollect_options} ${interval} ${ccollect_backups} | tee "${TEMP_LOG}"
build_backup_dir_list
move_log
send_report

View file

@ -1,65 +0,0 @@
#!/bin/sh
#
# -------------------------------------------------------------
# Get the real disk usage for a group of selected files
#
# This script counts the size of the files and directories
# listed, but exclude files that have hard links referenced outside
# the list.
#
# The undelying objective of this script is to report the
# real amount of disk used for backup solutions that are heavily
# using hard links to save disk space on identical files (I use
# ccollect, but this likely works with rsnapshot)
# -------------------------------------------------------------
# 20091002 - initial release - pdrolet (rdu@drolet.name)
# --------------------
# Parse options
# --------------------
# Known problem:
# - Command line cannot get a directory with a space in it
#
kdivider=1
find_options=""
while [ "$#" -ge 1 ]; do
case "$1" in
-m)
kdivider=1024
;;
-g)
kdivider=1048576
;;
-h|--help)
echo
echo $0: \<directories\> \[options below and any \"find\" options\]
echo \ \ -m: result in mega bytes \(rounded up\)
echo \ \ -g: result in giga bytes \(rounded up\)
echo \ \ -h: this help
echo
exit 0
;;
*)
find_options="${find_options} $1"
;;
esac
shift
done
# ------------------------------------------------------------------------------------------------------
# Compute the size
# ------------------------------------------------------------------------------------------------------
# 1) Find selected files and list link count, inodes, file type and size
# 2) Sort (sorts on inodes since link count is constant per inode)
# 3) Merge duplicates using uniq
# (result is occurence count, link count, inode, file type and size)
# 4) Use awk to sum up the file size of each inodes when the occurence count
# and link count are the same. Use %k for size since awk's printf is 32 bits
# 5) Present the result with additional dividers based on command line parameters
#
echo $((( `find ${find_options} -printf '%n %i %y %k \n' \
| sort -n \
| uniq -c \
| awk '{ if (( $1 == $2 ) || ($4 == "d")) { sum += $5; } } END { printf "%u\n",(sum); }'` \
+ ${kdivider} -1 ) / ${kdivider} ))

View file

@ -1 +0,0 @@
/var/cache/apt/archives/*

View file

@ -1,72 +0,0 @@
#!/bin/bash
# Backup-Ordner
BACKUP_DIR="/mnt"
# ccollect_logwrapper-Skript
CCOLLECT_LOGWRAPPER="./ccollect_logwrapper.sh"
# letzte Sicherung für Gruppe daily, weekly und monthly in Backup-Ordner ermitteln
DATE_DAILY=` ls $BACKUP_DIR | grep daily | sort -r | sed -e'2,$d' | cut -f 2 -d.`
DATE_WEEKLY=` ls $BACKUP_DIR | grep weekly | sort -r | sed -e'2,$d' | cut -f 2 -d.`
DATE_MONTHLY=`ls $BACKUP_DIR | grep monthly | sort -r | sed -e'2,$d' | cut -f 2 -d.`
DATE_YEARLY=` ls $BACKUP_DIR | grep yearly | sort -r | sed -e'2,$d' | cut -f 2 -d.`
# Falls Leerstring diesen mit "altem Datum" füllen
if [ -z "$DATE_DAILY" ] ; then DATE_DAILY="20000101-0101" ; fi
if [ -z "$DATE_WEEKLY" ] ; then DATE_WEEKLY="20000101-0101" ; fi
if [ -z "$DATE_MONTHLY" ] ; then DATE_MONTHLY="20000101-0101" ; fi
if [ -z "$DATE_YEARLY" ] ; then DATE_YEARLY="20000101-0101" ; fi
echo current: $DATE_CUR
echo last daily: $DATE_DAILY
echo last weekly: $DATE_WEEKLY
echo last monthly: $DATE_MONTHLY
echo last yearly: $DATE_YEARLY
# Datum date-konform wandeln
# Achtung: mit bash - nicht mit sh möglich!
# Alternativ mit expr... konvertieren
DATE_DAILY=${DATE_DAILY:0:4}-${DATE_DAILY:4:2}-${DATE_DAILY:6:2}" "${DATE_DAILY:9:2}:${DATE_DAILY:11:2}:00
DATE_WEEKLY=${DATE_WEEKLY:0:4}-${DATE_WEEKLY:4:2}-${DATE_WEEKLY:6:2}" "${DATE_WEEKLY:9:2}:${DATE_WEEKLY:11:2}:00
DATE_MONTHLY=${DATE_MONTHLY:0:4}-${DATE_MONTHLY:4:2}-${DATE_MONTHLY:6:2}" "${DATE_MONTHLY:9:2}:${DATE_MONTHLY:11:2}:00
DATE_YEARLY=${DATE_YEARLY:0:4}-${DATE_YEARLY:4:2}-${DATE_YEARLY:6:2}" "${DATE_YEARLY:9:2}:${DATE_YEARLY:11:2}:00
DATE_CUR=`date "+%Y-%m-%d %T"`
# Bei Bedarf Backups durchführen
if [ `date --date "$DATE_YEARLY" +%Y` -ne `date --date "$DATE_CUR" +%Y` ]
then
# Jahresbackup erzeugen
echo monthly backup started
source $CCOLLECT_LOGWRAPPER -a yearly
elif [ `date --date "$DATE_MONTHLY" +%Y%m` -ne `date --date "$DATE_CUR" +%Y%m` ]
then
# Monatsbackup erzeugen
echo monthly backup started
source $CCOLLECT_LOGWRAPPER -a monthly
elif [ `date --date "$DATE_WEEKLY" +%Y%W` -ne `date --date "$DATE_CUR" +%Y%W` ]
then
# Wochenbackup erzeugen
echo weekly backup started
source $CCOLLECT_LOGWRAPPER -a weekly
elif [ `date --date "$DATE_DAILY" +%Y%j` -ne `date --date "$DATE_CUR" +%Y%j` ]
then
# Tagesbackup erzeugen
echo daily backup started
source $CCOLLECT_LOGWRAPPER -a daily
else
# nichts zu tun
echo nothing to do
fi

View file

@ -0,0 +1,35 @@
to Local to Remote
backup destination is exiting
pre/postexec runs locally
--link-dest?
/delete_incomplete - can chech ddir
can check destination dir
-> dooooooo it before!
remote_host!
=> rddir_ls:
incomplete: ls -1 "${INTERVAL}"*".${c_marker}"
host support?
ssh-host-support?
=> ssh_host => save to host
execute commands there!
rm!
--link-dest?
--link-dest=DIR
=> remote dirs, rsync remote
=> works!!!!
local_destination
remote_destination
=> remote_*
both
configuration is local (what to where)

1
doc/braindumps/README Normal file
View file

@ -0,0 +1 @@
Do not read the files in this directory

View file

@ -30,7 +30,7 @@ Boot the system to be rescued from a media that contains low level tools
for your OS (like partitioning, formatting) and the necessary tools
(ssh, tar or rsync).
Use
- create the necessary volumes (like partitions, slices, ...)
- create the necessary partition table (or however it is called
Get a live-cd, that ships with
- rsync / tar

View file

@ -1,7 +1,7 @@
ccollect - Installing, Configuring and Using
============================================
Nico Schottelius <nico-ccollect__@__schottelius.org>
2.10, for ccollect 2.10, Initial Version from 2006-01-13
0.8, for ccollect 0.8, Initial Version from 2006-01-13
:Author Initials: NS
@ -21,12 +21,11 @@ Supported and tested operating systems and architectures
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
`ccollect` was successfully tested on the following platforms:
- GNU/Linux on amd64/hppa/i386/ppc/ARM
- FreeBSD on amd64/i386
- GNU/Linux on amd64/arm/hppa/i386/ppc
- Mac OS X 10.5
- Mac OS X 10.5
- NetBSD on alpha/amd64/i386/sparc/sparc64
- OpenBSD on amd64
- Windows by installing Cygwin, OpenSSH and rsync
It *should* run on any Unix that supports `rsync` and has a POSIX-compatible
bourne shell. If your platform is not listed above and you have it successfully
@ -73,19 +72,8 @@ Incompatibilities and changes
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Versions 0.9 and 1.0
^^^^^^^^^^^^^^^^^^^^
- Added "Error: " prefix in _exit_err()
Versions 0.8 and 0.9
^^^^^^^^^^^^^^^^^^^^
- Renamed script to ccollect (.sh is not needed)
- Removed feature to backup to a host via ccollect, added new tool
(FIXME: insert name here) that takes care of this via tunnel
- Perhaps creating subdirectory of source name (idea from Stefan Schlörholz)
Versions 0.7 and 0.8
^^^^^^^^^^^^^^^^^^^^
^^^^^^^^^^^^^^^^^^^^^
.The argument order changed:
- Old: "<interval name> [args] <sources to backup>"
@ -103,7 +91,7 @@ change for you.
- New: Options in $CCOLLECT_CONF/defaults are used as defaults (see below)
Versions 0.6 and 0.7
^^^^^^^^^^^^^^^^^^^^
^^^^^^^^^^^^^^^^^^^^^
.The format of `destination` changed:
- Before 0.7 it was a (link to a) directory
- As of 0.7 it is a textfile containing the destination
@ -115,7 +103,7 @@ You can update your configuration using `tools/config-pre-0.7-to-0.7.sh`.
Versions 0.5 and 0.6
^^^^^^^^^^^^^^^^^^^^
^^^^^^^^^^^^^^^^^^^^^
.The format of `rsync_options` changed:
- Before 0.6 it was whitespace delimeted
- As of 0.6 it is newline seperated (so you can pass whitespaces to `rsync`)
@ -132,7 +120,7 @@ XXXXX (- comes before digit).
Versions 0.4 and 0.5
^^^^^^^^^^^^^^^^^^^^
^^^^^^^^^^^^^^^^^^^^^
Not a real incompatibilty, but seems to fit in this section:
.0.5 does *NOT* require
@ -171,7 +159,7 @@ For those who do not want to read the whole long document:
--------------------------------------------------------------------------------
# get latest ccollect tarball from http://www.nico.schottelius.org/software/ccollect/
# replace value for CCV with the current version
export CCV=0.8.1
export CCV=0.8
#
# replace 'wget' with 'fetch' on bsd
@ -276,10 +264,8 @@ After having installed and used ccollect, report success using
Configuring
-----------
For configuration aid have a look at the above mentioned tools, which can assist
you quite well. When you are successfully using `ccollect`, I would be happy if
you add a link to your website, stating "I backup with ccollect", which points
to the ccollect homepage. So more people now about ccollect, use it and
improve it. You can also report success using `tools/report_success.sh`.
you quite well. When you are successfully using `ccollect`, report success using
`tools/report_success.sh`.
Runtime options
@ -347,15 +333,6 @@ If you add '$CCOLLECT_CONF/defaults/`pre_exec`' or
will start `pre_exec` before the whole backup process and
`post_exec` after backup of all sources is done.
If `pre_exec` exits with a non-zero return code, the whole backup
process will be aborted.
The `pre_exec` and `post_exec` script can access the following exported variables:
- 'INTERVAL': the interval selected (`daily`)
- 'no_sources': number of sources to backup (`2`)
- 'source_$no': name of the source, '$no' starts at 0 (`$source_0`)
The following example describes how to report free disk space in
human readable format before and after the whole backup process:
-------------------------------------------------------------------------
@ -620,16 +597,6 @@ respectively after doing the backup for *this specific* source.
If you want to have pre-/post-exec before and after *all*
backups, see above for general configuration.
If `pre_exec` exits with a non-zero return code, the backup
process of `this source` will be aborted (i.e. backup skipped).
The `post_exec` script can access the following exported variables from
ccollect:
- 'name': name of the source that is being backed up
- 'destination_name': contains the base directory name (`daily.20091031-1013.24496`)
- 'destination_dir': full path (`/tmp/ccollect/daily.20091031-1013.24496`)
- 'destination_full': like 'destination_dir', but prepended with the remote_host, if set (`host:/tmp/ccollect/daily.20091031-1013.24496` or `/tmp/ccollect/daily.20091031-1013.24496`)
Example:
--------------------------------------------------------------------------------
@ -653,22 +620,20 @@ If you create the file `delete_incomplete` in a source specification directory,
was interrupted) and remove them. Without this file `ccollect` will only warn
the user.
Detailed description of "rsync_failure_codes"
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
If you have the file `rsync_failure_codes` in your source configuration
directory, it should contain a newline-separated list of numbers representing
rsync exit codes. If rsync exits with any code in this list, a marker will
rsync exit codes. If rsync exits with any code in this list, a marker will
be left in the destination directory indicating failure of this backup. If
you have enabled delete_incomplete, then this backup will be deleted during
the next ccollect run on the same interval.
Detailed description of "mtime"
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
By default, ccollect.sh chooses the most recent backup directory for cloning or
the oldest for deletion based on the directory's last change time (ctime).
With this option, the sorting is done based on modification time (mtime). With
With this option, the sorting is done based on modification time (mtime). With
this version of ccollect, the ctime and mtime of your backups will normally
be the same and this option has no effect. However, if you, for example, move
your backups to another hard disk using cp -a or rsync -a, you should use this
@ -677,7 +642,6 @@ option because the ctimes are not preserved during such operations.
If you have any backups in your repository made with ccollect version 0.7.1 or
earlier, do not use this option.
Detailed description of "quiet_if_down"
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
By default, ccollect.sh emits a series of error messages if a source is not
@ -735,16 +699,12 @@ Using source names or interval in pre_/post_exec scripts
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The pre-/post_exec scripts can access some internal variables from `ccollect`:
- 'INTERVAL': The interval specified on the command line
- 'no_sources': number of sources
- 'source_$NUM': the name of the source
- 'name': the name of the currently being backuped source (not available for
- INTERVAL: The interval specified on the command line
- no_sources: number of sources
- source_$NUM: the name of the source
- name: the name of the currently being backuped source (not available for
generic pre_exec script)
Only available for `post_exec`:
- 'remote_host': name of host we backup to (empty if unused)
Using rsync protocol without ssh
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@ -865,7 +825,7 @@ If you want to see what changed between two backups, you can use
[12:00] u0255:ddba034.netstream.ch# rsync -n -a --delete --stats --progress daily.20080324-0313.17841/ daily.20080325-0313.31148/
--------------------------------------------------------------------------------
This results in a listing of changes. Because we pass -n to rsync no transfer
is made (i.e. report only mode).
is made (i.e. report only mode)"
This hint was reported by Daniel Aubry.
@ -983,31 +943,6 @@ you can enter your password (have a look at screen(1), especially "C-a M"
and "C-a _", for more information).
Backup fails, if autofs is running, but sources not reachable
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
If you are trying to backup a system containing paths that are managed
by autofs, you may run into this error:
-------------------------------------------------------------------------------
2009-12-01-23:14:15: ccollect 0.8.1: Beginning backup using interval monatlich
[ikn] 2009-12-01-23:14:15: Beginning to backup
[ikn] 2009-12-01-23:14:15: Executing /home/users/nico/ethz/ccollect/sources/ikn/pre_exec ...
Enter LUKS passphrase:
[ikn] Command successful.
[ikn] key slot 0 unlocked.
[ikn] 2009-12-01-23:14:23: Finished /home/users/nico/ethz/ccollect/sources/ikn/pre_exec (return code 0). [ikn] directory has vanished: "/home/users/nico/privat/firmen/ethz/autofs/projects"
[ikn] directory has vanished: "/home/users/nico/privat/firmen/ethz/autofs/scratch"
[ikn] directory has vanished: "/home/users/nico/privat/firmen/ethz/autofs/sgscratch"
[ikn] directory has vanished: "/home/users/nico/privat/firmen/ethz/autofs/supp"
[ikn] directory has vanished: "/home/users/nico/privat/firmen/ethz/autofs/sysadmin"
[ikn] rsync warning: some files vanished before they could be transferred (code 24) at main.c(1057) [sender=3.0.6]
[ikn] 2009-12-01-23:44:23: Source / is not readable. Skipping.
-------------------------------------------------------------------------------
Thus, if you are unsure whether autofs paths can be mounted during backup,
stop autofs in pre_exec and reenable it in post_exec.
Examples
--------
@ -1190,12 +1125,12 @@ rsync -av -H --delete /mnt/archiv/ "$DDIR/archiv/"
-------------------------------------------------------------------------
Processes running when doing ccollect -j
Processes running when doing ccollect -p
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Truncated output from `ps axuwwwf`:
-------------------------------------------------------------------------
S+ 11:40 0:00 | | | \_ /bin/sh /usr/local/bin/ccollect.sh daily -j ddba034 ddba045 ddba046 ddba047 ddba049 ddna010 ddna011
S+ 11:40 0:00 | | | \_ /bin/sh /usr/local/bin/ccollect.sh daily -p ddba034 ddba045 ddba046 ddba047 ddba049 ddna010 ddna011
S+ 11:40 0:00 | | | \_ /bin/sh /usr/local/bin/ccollect.sh daily ddba034
S+ 11:40 0:00 | | | | \_ /bin/sh /usr/local/bin/ccollect.sh daily ddba034
R+ 11:40 23:40 | | | | | \_ rsync -a --delete --numeric-ids --relative --delete-excluded --link-dest=/home/server/backup/ddba034

View file

@ -1,16 +0,0 @@
* Introduce -j option for max parallel jobs, deprecate -p (Darko Poljak)
* Add locking (Darko Poljak)
* Fix source-is-up check (Nikita Koshikov)
* Fix some minor command line parsing issues (Nico Schottelius)
* Correct output, if configuration is not in cconfig format (Nico Schottelius)
* Minor code cleanups and optimisations (Nico Schottelius)
* ccollect_analyse_logs.sh traps more errors and warnings (Patrick Drolet)
* Remove -v for mkdir and rm, as they are not POSIX (Patrick Drolet)
* Export destination_* to source specific post_exec (Nico Schottelius)
* Update documentation regarding exported variables (Nico Schottelius)
* Simplify time calculation (Nico Schottelius)
* Documentate pre_exec error handling (Nico Schottelius)
* Added start script (Thorsten Elle)
* Documentate autofs hint (Nico Schottelius)
* Speedup source-is-up check and remove --archive (Nico Schottelius)
* Removed support for remote backup (see doc) (Nico Schottelius)

View file

@ -1 +0,0 @@
* Add options for stdout, file and syslog logging (Darko Poljak)

View file

@ -1 +0,0 @@
* Add 'current' symlink to backup destinations (Steffen Zieger)

View file

@ -1 +0,0 @@
* Bugfix: empty rsync_options line causes destroying source (Darko Poljak)

View file

@ -1 +0,0 @@
* Bugfix: Fix parallel mode deadlock when max jobs > number of sources (Darko Poljak)

View file

@ -1,2 +0,0 @@
* Add Windows(Cygwin) as supported OS
* Add source name tag in log line

View file

@ -1 +0,0 @@
* Bugfix: exit in case of subshell error

View file

@ -1 +0,0 @@
* Improve performance, improve process of deletion of old backups

View file

@ -1 +0,0 @@
* Fix shellcheck reported issues

View file

@ -1 +0,0 @@
* Fix excluding destination dir from removal

View file

@ -1 +0,0 @@
* Make rsync return code available in post_exec (Steffen Zieger)

View file

@ -10,7 +10,7 @@ ccollect - (pseudo) incremental backup with different exclude lists using hardli
SYNOPSIS
--------
'ccollect.sh' [args] <interval name> <sources to backup>
'ccollect.sh' <interval name> [args] <sources to backup>
DESCRIPTION
@ -26,54 +26,17 @@ texinfo or html).
OPTIONS
-------
-a, --all::
Backup all sources specified in /etc/ccollect/sources
-e, --errors::
Log only errors
-h, --help::
Show the help screen
-j [max], --jobs [max]::
Specifies the number of jobs to run simultaneously.
If max is not specified then parallelise all jobs.
-l FILE, --logfile FILE::
Log to specified file
-p, --parallel::
Parallelise backup processes (deprecated from 2.0)
Parallelise backup processes
-s, --syslog::
Log to syslog with tag ccollect
-V, --version::
Show version and exit
-a, --all::
Backup all sources specified in /etc/ccollect/sources
-v, --verbose::
Be very verbose (uses set -x)
LOGGING MECHANISM
-----------------
ccollect logging depends on running in non-interactive/interactive mode
and on specified optins. The mechanism behaves as the following:
non-interactive mode::
* standard output goes to syslog
* optional: specify logging into file
* log all output by default
* optional: log only errors
interactive mode::
* standard output goes to stdout
* log only errors
* optional: log into syslog or file
- log all output by default
- optional: log only errors
Be very verbose (uses set -x).
SEE ALSO

View file

@ -1,7 +1,8 @@
* Change version and date in ccollect.sh
* Change version in documentation/ccollect.text
* Regenerate documentation
* Create tarball
* Transfer to website
* Transfer to home.schottelius.org
* Extract files
* Update website
* Announce on freshmeat

4
doc/todo/0.5.2 Normal file
View file

@ -0,0 +1,4 @@
x Fix $? problem
x Check last dir searching
x Check count generation
x Check general functionality (remove testing)

34
doc/todo/0.5.3 Normal file
View file

@ -0,0 +1,34 @@
Done:
==> screenshot
u0219 zrha166.netstream.ch # ~chdscni9/ccollect.sh weekly zrha166.netstream.ch
2007-08-16-21:49:44: ccollect 0.6: Beginning backup using interval weekly
[zrha166.netstream.ch] 2007-08-16-21:49:44: Beginning to backup
[zrha166.netstream.ch] 2007-08-16-21:49:45: Existing backups: 0 Total keeping backups: 8
[zrha166.netstream.ch] 2007-08-16-21:49:45: Did not find existing backups for interval weekly.
[zrha166.netstream.ch] 2007-08-16-21:49:45: Using backup from daily.
[zrha166.netstream.ch] 2007-08-16-21:49:45: Beginning to backup, this may take some time...
[zrha166.netstream.ch] 2007-08-16-21:49:45: Creating /etc/ccollect/sources/zrha166.netstream.ch/destination/weekly.20070816-2149.22188 ...
[zrha166.netstream.ch] 2007-08-16-21:49:45: Transferring files...
- beep-after-delete-hack?
-> tonnerre / #cLinux
- replace nico-linux-ccollect with nico-ccollect
* ccollect is not Linux specific
- remove exit-calls
* will leave behind unused temporary file!
* use _exit_err
X join todos
- fix possible quoting problems
* rsync_extra (redefine format)
* exclude
* other
* create _rsync, filtering args, creating new $@
- check and export variables for use in scripts!
Contact Julian later:
* integrate updated german documentation

11
doc/todo/0.5.3.tonnerre Normal file
View file

@ -0,0 +1,11 @@
NetBSD/i386,amd64,sparc,sparc64
13:13 < Tonnerre> telmich, die kleine wä, 2 Variablen aus $(CCOLLECT) zu
machen
13:13 < Tonnerre> telmich, eine fü Sourcepfad und eine fü
Destinationfpad
13:13 < Tonnerre> pfa
13:13 < Tonnerre> d
13:14 < Tonnerre> telmich, die gröre wä die $() durch ${} zu ersetzen, so
dass der Kram auch mit !GNU-Make geht

63
doc/todo/0.6 Normal file
View file

@ -0,0 +1,63 @@
not did, did not remember why I wanted to do that:
- do sed-time check:
1.2. replace sed?
compare timing:
_echo () { echo "$name $msg" }
and create
_techo () { echo "$timestamp $name $msg" }
perhaps create
_eecho () { _techo "ERROR $msg" }
?
done:
add note for
09:24 < telmich> Obri: ~/.ssh/config!
- implement use of different intervals as source for cloning
* use 'woechentlich' if available and no backup exists for 'daily'
* add 'prefer_latest' to prefer different interval, that's newer than
ours
* or add 'prefer_same_interval' instead?
- implement detection of partial backups
3. detect unfinished backups
sven / markierung
- wie seht der Marker aus?
-> .ccollect-YEAR-MM-DD.HHmm.pid (like the directory)
--> assume incomplete, when we did not finish.
--> means it's complete,when rsync was successful
---> partial implemented in 0.5.2 (commented out)
- do not check by name, but by time
* is ls -t posix?
* also check: -p -1
- ccollect Zeitausgabe verbessern
- Wofuer wie lange gebraucht
* rm
* klonen (gibt es so ja nicht)
Wenn Summary angegeben ist am Ende ausgeben
- do we want rsync -H by default?
* no: takes much more memory
ssh / port change:
change rsync_extra format
- Variables:
source_$n
no_sources
name
- changed naming of sources:
YYYYMMDD-HHMM.PID (better readable)
=> source / backup converter! => not needed!
config:
set destination-base
/etc/ccollect/defaults/source_config
Documentation:
- hint: backuping backup
- update documentation:
- german doc?
- exit pre/post exec -> error codes (after implementation!) (in 0.4)
- write about fast changing fs
- delete_incomplete

23
doc/todo/0.6.1 Normal file
View file

@ -0,0 +1,23 @@
- Add filter support
* filter
- add source/ignore_failed_pre_exec
- add source/ignore_failed_post_exec
- do not delete_incomplete, when there's only one backup left!
- .ccollect-marker is deleted by rsync at the beginning!
- fix marking
- add logwrapper
- add loganalyser
speedup is
error codes
vanished files (see netstream)!!!
Done:
- Improve finding backup from another interval:
o strip of interval name
o run sort -n
o use the last entry
- add --version, -V
not needed:
- think about 'prefer_same_interval' / 'prefer_latest'

1
doc/todo/0.6.2 Normal file
View file

@ -0,0 +1 @@
- fix delete_incomplete marker

35
doc/todo/0.7.1 Normal file
View file

@ -0,0 +1,35 @@
--------------------------------------------------------------------------------
done:
- tools in makefile
14:15 <telmich> also alle tools/ccollect_*
14:15 <dev-zero> mach mal n besseres Makefile :)
14:15 <telmich> hatte die extra deswegen umbenannt
14:15 <telmich> nehme ich mal als hinweis für 0.7.1
- add global delete_incomplete (/etc/ccollect/defaults/delete_incomplete)
09:31 < Obri> telmich: hab nen kleinen tipp für ccollect
09:32 < Obri> telmich: ich habe hier hosts die nicht immer online sind, das ist dumm weil so das backup
kaputtgeht...
09:32 < Obri> telmich: ich habe jetzt das ein preexec script gebastelt:
09:32 < Obri> ping -c1 -q `cat /etc/ccollect/sources/$name/source | cut -d"@" -f2 | cut -d":" -f1`
09:33 < Obri> telmich: so bricht das backup ab wenn der host nicht erreichbar ist
09:33 < Obri> ohne dass ein altes backup entsorgt wird
- remove basename
-> include config from cconf! -> standard!
reject:
--------------------------------------------------------------------------------
- fix german documentation!
=> I'm the coder, somebody else can fix it.
- add wrapper, to do logging and analyse
* output filename in the wrapper, save into variable
=> mktemp
* call analyser
=> output stdout?
=> no use for that currently

63
doc/todo/0.7.2 Normal file
View file

@ -0,0 +1,63 @@
--------------------------------------------------------------------------------
Stats version
--------------------------------------------------------------------------------
Add tools/ccollect_stats.sh, clearify license
--------------------------------------------------------------------------------
Preamble:
Netstream (www.netstream.ch) may consider using ccollect for backing up many
unix servers and needs some clean reporting. The following things
need to be done, so ccollect will be useful for netstream:
Logger:
- Needs to write small mails (sysadmins don't have time to read mails)
- Needs to be able to only write a mail on error
* needs configuration option to also mail on warnings
- Should be able to send one mail per backup source
* or one for the whole backup job
Messages: (to be used for filtering)
Errors:
Read from remote host .*: Connection timed out
Warnings:
rsync: mknod .* failed: Invalid argument (22)
file has vanished: ".*"
--------------------------------------------------------------------------------
Analyzer:
- grosse Dateien
- grosse Veraenderungen
--------------------------------------------------------------------------------
exclude-lists-doku:
freebsd:
/usr/ports/*
/proc/*
/dev/*
/tmp/*
/var/tmp/*
linux:
/sys/*
/proc/*
/dev/*
/tmp/*
--------------------------------------------------------------------------------
done:
rsync_extra global!
- \n delimeted
--------------------------------------------------------------------------------
-S, --sparse
Try to handle sparse files efficiently so they take up less space on the des
tination. Conflicts with --inplace because its not possible to overwrite
data in a sparse fashion.
--------------------------------------------------------------------------------
Always report return code!
[12:00] u0255:ddba034.netstream.ch# rsync -n -a --delete --stats --progress daily.20080324-0313.17841/ daily.20080325-0313.31148/
$tool <name> <version1> <version2>

2
doc/todo/0.7.3 Normal file
View file

@ -0,0 +1,2 @@
- add -a (archive) to ccollect_delete_source.text
- add listing of intervals to ccollect_list_intervals

1
doc/todo/0.7.4 Normal file
View file

@ -0,0 +1 @@
add support for wipe

6
doc/todo/0.8.0 Normal file
View file

@ -0,0 +1,6 @@
- restructure code to easily allow global versus source options:
f_name=....
check_option $f_name
=> source
=> defaults
- support all senseful options as default and source specific

View file

@ -1,12 +1,6 @@
ccollect todos
==============
Nico Schottelius <nico-ccollect__@__schottelius.org>
0.1, for the current ccollect version
:Author Initials: NS
These todos are things that would be senseful todo, but are just nice
to have. This means I won't code them, so somebody can code them.
--------------------------------------------------------------------------------
- Add ccollect-restore.sh
(new project, perhaps coordinated with jchome's ccollect-config)
@ -30,16 +24,3 @@ to have. This means I won't code them, so somebody can code them.
- add option ("run_only_once") to prevent ccollect to run twice
(per source/general)
-> or create wrapper, that takes care of it
- add support for wipe
- cleanup doc/: remove old stuff, make clear what can be installeld,
cleanup ccollect-restoring.text
- clean logwrapper.text and probably create logwrapper, that does
what is being stated in logwrapper.text
- include pdf support again (Makefile:72)
- integrate website into ccollect -> updates are done here

View file

@ -1 +0,0 @@
5

View file

@ -1 +0,0 @@
4

View file

@ -1 +0,0 @@
4

View file

@ -1 +0,0 @@
2

View file

@ -1,3 +0,0 @@
#!/bin/sh
echo 'General post_exec executed.'

View file

@ -1,3 +0,0 @@
#!/bin/sh
echo 'General pre__exec executed.'

View file

@ -1 +0,0 @@
/tmp/ccollect/backup

View file

@ -1 +0,0 @@
.git

View file

@ -1 +0,0 @@
/tmp/ccollect/source

View file

@ -1 +0,0 @@
/tmp/ccollect/backup

View file

@ -1 +0,0 @@
.git

View file

@ -1 +0,0 @@
/tmp/ccollect/source

View file

@ -1 +0,0 @@
/tmp/ccollect/backup-chint

View file

@ -1 +0,0 @@
.git

View file

@ -1 +0,0 @@
/tmp/ccollect/source

View file

@ -1 +0,0 @@
/tmp/ccollect/backup

View file

@ -1 +0,0 @@
.git

View file

@ -1 +0,0 @@
/tmp/ccollect/source

View file

@ -1 +0,0 @@
/tmp/ccollect/backup

View file

@ -1 +0,0 @@
/tmp/ccollect/source

View file

@ -1 +0,0 @@
/tmp/ccollect/backup

View file

@ -1 +0,0 @@
.git

View file

@ -1 +0,0 @@
/tmp/ccollect/source

View file

@ -1 +0,0 @@
/tmp/ccollect/backup

View file

@ -1,5 +0,0 @@
#!/bin/cat
######################################################################
Source post_exec executed.
######################################################################

View file

@ -1,5 +0,0 @@
#!/bin/cat
######################################################################
Source pre_exec executed.
######################################################################

View file

@ -1 +0,0 @@
/tmp/ccollect/source

18
test/exec.sh Executable file
View file

@ -0,0 +1,18 @@
#!/bin/sh
host="home.schottelius.org"
host=""
set -x
pcmd()
{
echo "$#", "$@"
if [ "$host" ]; then
ssh "$host" "$@"
else
$@
fi
}
#pcmd ls /
#pcmd cd /; ls "/is not there"
pcmd cd / && ls

1
test/local.sh Executable file
View file

@ -0,0 +1 @@
CCOLLECT_CONF=./conf ./ccollect.sh daily -v local1

1
test/remote.sh Executable file
View file

@ -0,0 +1 @@
CCOLLECT_CONF=./conf ./ccollect.sh daily -v remote1

23
test/return-value.sh Executable file
View file

@ -0,0 +1,23 @@
#!/bin/sh
ls /surely-not-existent$$ 2>/dev/null
if [ "$?" -ne 0 ]; then
echo "$?"
fi
ls /surely-not-existent$$ 2>/dev/null
ret=$?
if [ "$ret" -ne 0 ]; then
echo "$ret"
fi
# if is true, ls is fales
if [ "foo" = "foo" ]; then
ls /surely-not-existent$$ 2>/dev/null
fi
# but that's still the return of ls and not of fi
echo $?

23
tools/ccollect_push → test/test-ccollect-tools.sh Executable file → Normal file
View file

@ -1,30 +1,29 @@
#!/bin/sh
#
# 2011 Nico Schottelius (nico-ccollect at schottelius.org)
#
#
# 2009 Nico Schottelius (nico-ccollect at schottelius.org)
#
# This file is part of ccollect.
#
# ccollect is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
#
# ccollect is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
#
# You should have received a copy of the GNU General Public License
# along with ccollect. If not, see <http://www.gnu.org/licenses/>.
#
# Support backup triggering and actual backup through
# ssh tunnel
#
# Use the supplied called_from_remote script as pre…exec
#
# Test the ccollect tools suite
#
export ssh_remote_port="$1"; shift
export backup_host="$1"; shift
set -x
ssh -R$ssh_remote_port:127.0.0.1:22 "$backup_host" "ccollect" "$@"
tmp="$(mktemp /tmp/ccollect-tools.XXXXXXXXXXX)"
rm -rf "${tmp}"

44
test/test-ccollect1.sh Executable file
View file

@ -0,0 +1,44 @@
#!/bin/sh
#
# Nico Schottelius <nico-linux //@// schottelius.org>
# Date: 27-Jan-2007
# Last Modified: -
# Description:
#
ccollect=../ccollect.sh
testdir="$(dirname $0)/test-backups"
confdir="$(dirname $0)/test-config"
source="$(hostname)"
source_source="/tmp"
interval="taeglich"
# backup destination
mkdir -p "$testdir"
source_dest="$(cd "$testdir"; pwd -P)"
# configuration
mkdir -p "${confdir}/sources/${source}"
ln -s "$source_dest" "${confdir}/sources/${source}/destination"
echo "$source_source" > "${confdir}/sources/${source}/source"
touch "${confdir}/sources/${source}/summary"
touch "${confdir}/sources/${source}/verbose"
mkdir -p "${confdir}/defaults/intervals/"
echo 3 > "${confdir}/defaults/intervals/$interval"
# create backups
CCOLLECT_CONF="$confdir" "$ccollect" "$interval" -p -a
touch "${source_source}/$(date +%s)-$$.1982"
CCOLLECT_CONF="$confdir" "$ccollect" "$interval" -p -a
touch "${source_source}/$(date +%s)-$$.42"
CCOLLECT_CONF="$confdir" "$ccollect" "$interval" -p -a
du -sh "$testdir"
du -shl "$testdir"
echo "Delete $testdir and $confdir after test"

View file

@ -1,18 +1,19 @@
Files and their tasks / destinations:
ccollect_add_source: bin
ccollect_analyse_logs: bin
ccollect_create_source2: ???
ccollect_create_source: ???
ccollect_delete_source: bin
ccollect_list_intervals: bin
ccollect_logwrapper: bin
ccollect-stats: ???
ccollect_add_source.sh: bin
ccollect_analyse_logs.sh: bin
ccollect_create_source2.sh: ???
ccollect_create_source.sh: ???
ccollect_delete_source.sh: bin
ccollect_list_intervals.sh: bin
ccollect_logwrapper.sh: bin
ccollect-stats.sh: ???
config-pre-0.4-to-0.4.BUGS: only to be used for converting
config-pre-0.4-to-0.4: only to be used for converting
config-pre-0.4-to-0.4.sub: only to be used for converting
config-pre-0.6-to-0.6: only to be used for converting
config-pre-0.6-to-0.6.sub: only to be used for converting
config-pre-0.7-to-0.7: only to be used for converting
config-pre-0.7-to-0.7.sub: only to be used for converting
config-pre-0.4-to-0.4.sh: only to be used for converting
config-pre-0.4-to-0.4.sub.sh: only to be used for converting
config-pre-0.6-to-0.6.sh: only to be used for converting
config-pre-0.6-to-0.6.sub.sh: only to be used for converting
config-pre-0.7-to-0.7.sh: only to be used for converting
config-pre-0.7-to-0.7.sub.sh: only to be used for converting
gnu-du-backup-size-compare.sh
README

View file

@ -1,30 +0,0 @@
#!/bin/sh
#
# 2011 Nico Schottelius (nico-ccollect at schottelius.org)
#
# This file is part of ccollect.
#
# ccollect is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# ccollect is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with ccollect. If not, see <http://www.gnu.org/licenses/>.
#
# Support backup triggering and actual backup through
# ssh tunnel
#
# Use the supplied called_from_remote script as pre…exec
#
#
export ssh_remote_port="$1"; shift
export backup_host="$1"; shift
ssh -R$ssh_remote_port:127.0.0.1:22 "$backup_host" "ccollect" "$@"

View file

@ -72,11 +72,6 @@ if [ "$search_err" ]; then
set -- "$@" "-e" 'ssh: connect to host .*: Connection timed out'
set -- "$@" "-e" 'rsync error: unexplained error (code 255)'
set -- "$@" "-e" 'rsync: connection unexpectedly closed'
set -- "$@" "-e" 'rsync: send_files failed to open'
set -- "$@" "-e" 'rsync: readlink_stat .* failed: File name too long'
set -- "$@" "-e" 'IO error encountered .* skipping file deletion'
set -- "$@" "-e" 'rsync: read error: Connection reset by peer'
set -- "$@" "-e" 'rsync error: error in rsync protocol data stream'
fi
# known error strings:
@ -108,7 +103,6 @@ fi
if [ "$search_warn" ]; then
# warn on non-zero exit code
set -- "$@" "-e" 'Finished backup (rsync return code: [^0]'
set -- "$@" "-e" 'rsync: file has vanished'
set -- "$@" "-e" 'WARNING: .* failed verification -- update discarded (will try again).'
fi
# known warnings:
@ -119,12 +113,9 @@ fi
# Interesting strings in the logs: informational
# ----------------------------------------------
if [ "$search_info" ]; then
set -- "$@" "-e" 'Number of files: [[:digit:]]*'
set -- "$@" "-e" 'Number of files transferred: [[:digit:]]*'
set -- "$@" "-e" 'Total transferred file size: [[:digit:]]* bytes'
set -- "$@" "-e" 'total size is [[:digit:]]* speedup is'
set -- "$@" "-e" 'Backup lasted: [[:digit:]]*:[[:digit:]]\{1,2\}:[[:digit:]]* (h:m:s)$'
set -- "$@" "-e" 'sent [[:digit:]]* bytes received [0-9]* bytes'
set -- "$@" "-e" 'send [[:digit:]]* bytes received [0-9]* bytes [0-9]* bytes/sec$'
fi
# info includes:
@ -132,6 +123,4 @@ fi
# [u0160.nshq.ch.netstream.com] 2007-08-20-18:26:06: Backup lasted: 0:43:34 (h:m:s)
#[ddba012.netstream.ch] sent 3303866 bytes received 1624630525 bytes 122700.92 bytes/sec
#echo Parameters: "$@"
grep "$@"

Some files were not shown because too many files have changed in this diff Show more