Setup backup server configuration for backup replication.

files/opt/backup/filter_backups.py

@@ -0,0 +1,117 @@
+from datetime import datetime
+from dateutil.relativedelta import relativedelta
+from argparse import ArgumentParser, FileType
+
+def filter_backups_timeslot(offset, backups, timeslot='days', keep=7, mode='max', printy=True):
+    keep_backups = []
+    for timeslot_off in range(1, keep):
+        indices = [
+            idx
+            for idx,value in enumerate(backups)
+            if (
+                value[1] >= (offset - relativedelta(**{timeslot:timeslot_off}))
+            )
+        ]
+        if not indices:
+            continue
+        if mode == "min":
+            #removed not considered backups for this slot!
+            for idx in reversed(indices[1:]):
+                backups.pop(idx)
+            keep_backups.append(backups.pop(indices[0]))
+        elif mode == "max":
+            keep_backups.append(backups.pop(indices[-1]))
+            #removed not considered backups for this slot!
+            for idx in reversed(indices[:-1]):
+                backups.pop(idx)
+        else:
+            for idx in sorted(indices, reverse=True):
+                keep_backups.append(backups.pop(idx))
+    return keep_backups, backups
+
+def filter_backups(backuplist, show_keep=True, keep_hours=24, keep_days=7, keep_weeks=4, keep_months=12, keep_years=10):
+    backups = sorted(
+        [
+            (backup, datetime.fromtimestamp(int(backup.split('-')[1])))
+            for backup in backuplist
+        ], key=lambda v: v[1], reverse=True
+    )
+    offset = backups[0][1]
+    keep_backups = [backups.pop(0)]
+
+    part_keep_backups, backups = filter_backups_timeslot(offset, backups, timeslot='hours',keep=keep_hours,mode='all')
+    keep_backups.extend(part_keep_backups)
+    part_keep_backups, backups = filter_backups_timeslot(offset, backups, timeslot='days',keep=keep_days,mode='max')
+    keep_backups.extend(part_keep_backups)
+    part_keep_backups, backups = filter_backups_timeslot(offset, backups, timeslot='weeks',keep=keep_weeks,mode='max')
+    keep_backups.extend(part_keep_backups)
+    part_keep_backups, backups = filter_backups_timeslot(offset, backups, timeslot='months',keep=keep_months,mode='max')
+    keep_backups.extend(part_keep_backups)
+    part_keep_backups, backups = filter_backups_timeslot(offset, backups, timeslot='years',keep=keep_years,mode='max')
+    keep_backups.extend(part_keep_backups)
+
+    keeps = [backup[0] for backup in keep_backups]
+    if not show_keep:
+        return [backup for backup in backuplist if backup not in keeps]
+    return keeps
+
+def main():
+    parser = ArgumentParser(description="Program to print backups to keep or delete based on time stamps.")
+    parser.add_argument(
+        "-f", "--file", metavar="FILE", type=FileType('r'), required=True,
+        help="Path to file with list of backups to filter. Expects per line one backup name in format <appname>-<timestamp>")
+    parser.add_argument(
+        "-d", "--print-delete", action="store_true", required=False,
+        help="Instead of objects to kept, print out objects to be deleted."
+    )
+    parser.add_argument(
+        "-H", "--hours-keep", metavar="KEEP", type=int, required=False,
+        default=24, help="Number of objects to keep on hourly basis."
+    )
+    parser.add_argument(
+        "-D", "--days-keep", metavar="KEEP", type=int, required=False,
+        default=7, help="Number of objects to keep on daily basis."
+    )
+    parser.add_argument(
+        "-W", "--weeks-keep", metavar="KEEP", type=int, required=False,
+        default=4, help="Number of objects to keep on weekly basis."
+    )
+    parser.add_argument(
+        "-M", "--months-keep", metavar="KEEP", type=int, required=False,
+        default=12, help="Number of objects to keep on monthly basis."
+    )
+    parser.add_argument(
+        "-Y", "--years-keep", metavar="KEEP", type=int, required=False,
+        default=10, help="Number of objects to keep on yearly basis."
+    )
+
+    args = parser.parse_args()
+    show_keep = not args.print_delete
+    keep_hours = args.hours_keep
+    keep_days = args.days_keep
+    keep_weeks = args.weeks_keep
+    keep_months = args.months_keep
+    keep_years = args.years_keep
+
+    backuplist = [
+        line.strip()
+        for line in args.file
+        if len(line.strip()) > 0
+    ]
+    fail = False
+    for idx,backup in enumerate(backuplist):
+        if len(backup) < 10 or not backup[-10:].isnumeric():
+            print(f"WARNING! Line {idx+1} has no timestamp suffix")
+            fail = True
+    if fail:
+        exit(1)
+
+    considered_backups = filter_backups(
+        backuplist, show_keep,
+        keep_hours, keep_days, keep_weeks, keep_months, keep_years
+    )
+    for backup in considered_backups:
+        print(backup)
+
+if __name__ == '__main__':
+    main()