commit 7f2db05ae1fe79503f975a350ec50b5a0d2c2808
Author: Jakobus Schürz <jakobus.schuerz@home.arpa>
Date:   Sun Sep 1 09:42:49 2024 +0200

    first commit

diff --git a/README.md b/README.md
new file mode 100644
index 0000000..7d237e5
--- /dev/null
+++ b/README.md
@@ -0,0 +1,22 @@
+#CLEANUP MATRIX SYNAPSE
+
+Das script per Cronjob täglich laufen lassen.
+
+Und von Zeit zu Zeit verbindet man sich mit der Postgresql-Datenbank als User postgres
+
+`sudo -u postgresql psql`
+
+Dann matrix-synapse stoppen und im Anschluss mit der Datenbank für synapse connecten
+
+```
+\c synapse
+REINDEX (VERBOSE) DATABSE synapse;
+VACUUM FULL VERBOSE;
+```
+
+Das reduziert die Größe der Datenbank enorm. 
+
+Anschließend wieder matrix-synapse starten.
+
+Quellen: 
+- https://levans.fr/shrink-synapse-database.html
diff --git a/cleanup_matrix.sh b/cleanup_matrix.sh
new file mode 100644
index 0000000..6acb1d6
--- /dev/null
+++ b/cleanup_matrix.sh
@@ -0,0 +1,129 @@
+#!/bin/bash
+
+# There are mostly three reasons the synapse database tends to get large over time:
+#
+#  - Stuff that no longer needs to be kept around and should be deleted
+#  - Synapse is extremely cache-happy, and this takes a lot of space
+#  - Table bloat & Index bloat in PostgreSQL
+#
+# See also:
+# https://github.com/matrix-org/synapse/wiki/SQL-for-analyzing-Synapse-PostgreSQL-database-stats
+
+# Exit on first failure
+set -e
+
+# Adjust as needed to your own setup
+TOKEN="<TOKEN from matrix-client"
+export PGPASSWORD="<PG_PASSWORD>"
+PGUSER="synapse_user"
+PGDB="synapse"
+PGHOST="localhost"
+FQDN="<your.matrix-server.fqdn>"
+
+WORKDIR=/root/matrix_cleanup
+SUJET="Matrix server cleanup - $(hostname -f)";
+POSTMASTER="root"
+RESULTAT="KO"
+LOGFILE=/var/log/matrix/cleanup.log
+
+# Ensure dirs exist
+[[ -d $(dirname ${LOGFILE}) ]] || mkdir -p $(dirname ${LOGFILE})
+[[ -d ${WORKDIR} ]] || mkdir -p ${WORKDIR}
+
+jnl() {
+    echo "$(date +%c) : $1" | tee -a "${LOGFILE}"
+}
+
+two_months_ago() {
+    timestamp=$(date +"%s" -d '2 months ago')
+    echo "${timestamp}"
+}
+
+one_year_ago() {
+    date +"%s%N" -d '1 year ago'|cut -b1-13
+}
+
+notify() {
+   jnl "End of programme."
+   cat $LOGFILE | mail -s "${SUJET} : ${RESULTAT}" ${POSTMASTER}
+   [ "${RESULTAT}" = "OK" ] && exit 0 || exit 1
+}
+
+#####################################
+#      Main
+#####################################
+
+# pre-cleanup
+jnl "Remove temporary files..."
+rm -f /opt/synapse-compress/*.sql
+rm -f /root/matrix_cleanup/*.txt
+
+jnl "Dumping current status on Database"
+
+biggestTables=$(echo "SELECT nspname || '.' || relname AS \"relation\", pg_size_pretty(pg_total_relation_size(C.oid)) AS "total_size" FROM pg_class C LEFT JOIN pg_namespace N ON (N.oid = C.relnamespace) WHERE nspname NOT IN ('pg_catalog', 'information_schema') AND C.relkind <> 'i' AND nspname !~ '^pg_toast' ORDER BY pg_total_relation_size(C.oid) DESC LIMIT 20;" | psql -h ${PGHOST} -d ${PGDB} -U ${PGUSER})
+
+jnl "${biggestTables}"
+
+# Removing empty rooms
+jnl "Getting empty rooms..."
+
+curl -H "HOST: ${FQDN}" -H "Authorization: Bearer $TOKEN" 'http://127.0.0.1:8008/_synapse/admin/v1/rooms?limit=300' | jq '.rooms[] | select(.joined_local_members == 0) | .room_id'|sed 's/^"//;s/"$//' > ${WORKDIR}/to_purge.txt
+
+jnl < ${WORKDIR}/to_purge.txt
+
+while read room_id; do 
+    jnl "purging room ${room_id}..."
+    #curl -w "\n" --header "Authorization: Bearer $TOKEN" -X POST -H "Content-Type: application/json" -d "{ \"room_id\": $room_id }" -H "HOST: ${FQDN}" 'http://127.0.0.1:8008/_synapse/admin/v1/purge_room'
+    curl -w "\n" --header "Authorization: Bearer $TOKEN" -X DELETE -H "Content-Type: application/json" -d "{ \"purge\": true }" -H "HOST: ${FQDN}" "http://127.0.0.1:8008/_synapse/admin/v2/rooms/${room_id}"
+     sleep 0.5
+done < ${WORKDIR}/to_purge.txt
+
+jnl "Deleting history of more than 2 months in large rooms"
+
+curl -H "HOST: ${FQDN}" -H "Authorization: Bearer $TOKEN" 'http://127.0.0.1:8008/_synapse/admin/v1/rooms?limit=300' | jq '.rooms[] | select(.state_events > 100) | .room_id' | sed 's/\"//g' > ${WORKDIR}/history_to_purge.txt
+
+if [[ -s ${WORKDIR}/history_to_purge.txt ]]; then
+    dateUntil=$(two_months_ago)
+
+    while read room_id; do 
+        jnl "purging history for ${room_id}..."
+        echo curl  -w "\n" --header "Authorization: Bearer $TOKEN" -X POST -H "Content-Type: application/json" -d "{ \"delete_local_events\": false, \"purge_up_to_ts\": ${dateUntil} }" -H "HOST: ${FQDN}" "http://127.0.0.1:8008/_synapse/admin/v1/purge_history/$room_id"
+        curl  -w "\n" --header "Authorization: Bearer $TOKEN" -X POST -H "Content-Type: application/json" -d "{ \"delete_local_events\": false, \"purge_up_to_ts\": ${dateUntil} }" -H "HOST: ${FQDN}" "http://127.0.0.1:8008/_synapse/admin/v1/purge_history/$room_id"
+        sleep 0.5
+    done < ${WORKDIR}/history_to_purge.txt 
+fi
+
+    echo curl -X POST -H "HOST: ${FQDN}" -H "Authorization: Bearer $TOKEN"  "http://127.0.0.1:8008/_synapse/admin/v1/purge_media_cache?before_ts=$(one_year_ago)"
+    curl -X POST -H "HOST: ${FQDN}" -H "Authorization: Bearer $TOKEN"  "http://127.0.0.1:8008/_synapse/admin/v1/purge_media_cache?before_ts=$(one_year_ago)"
+
+jnl "Optimizing synapse cache for rooms with more than 100 000 state changes"
+# Get the list of rooms with more than 99 999 state groups events
+
+jnl "Gathering list"
+psql -h "${PGHOST}" -d "${PGDB}" -U "${PGUSER}" -t -c "SELECT room_id, count(*) AS count FROM state_groups_state GROUP BY room_id HAVING count(*) > 99999 ORDER BY count DESC;" | sed -r 's/\s//g' | egrep -v '^$' |awk -F "|" '{print $1}' > ${WORKDIR}/to_compress.txt
+
+jnl "Prepare compression files state-compressor"
+cpt=1
+while read room_id; do 
+    # Could we remove the password / is it using .pgpass?
+    echo room_id ${room_id}
+    /usr/local/bin/synapse-compress-state -t -o /opt/synapse-compress/state-compressor_${cpt}.sql -p "host=${PGHOST} user=${PGUSER} password=${PGPASSWORD} dbname=${PGDB}" -r "$room_id"
+    ((cpt++))
+    sleep 0.5
+done < ${WORKDIR}/to_compress.txt
+
+jnl "Running the compressions..."
+for file in `ls /opt/synapse-compress/state-compressor*.sql`; do 
+    jnl "Compressing ${file}"
+    psql -h ${PGHOST} -d ${PGDB} -U ${PGUSER} < ${file}
+    sleep 0.5
+done
+
+# If needed, or to do once in a while 
+# REINDEX (VERBOSE) DATABASE matrix_prod;
+# VACUUM FULL VERBOSE state_groups_state;
+
+RESULTAT="OK"
+notify
+
+exit 0