1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
| #!/usr/bin/env python3
"""
Open WebUI SQLite -> PostgreSQL Migration Script
Usage:
python3 migrate_sqlite_to_pg.py \
--sqlite /path/to/webui.db \
--pg "postgresql://user:pwd@127.0.0.1:5432/postgres"
Prerequisites:
pip install psycopg2-binary
Notes:
- The target PostgreSQL database must already have the schema created
(by running Open WebUI once with the PG connection).
- This script will TRUNCATE all target tables before inserting.
- Run this with the Open WebUI container STOPPED.
"""
import argparse
import json
import sqlite3
import sys
import time
try:
import psycopg2
import psycopg2.extras
except ImportError:
print("ERROR: psycopg2 not installed. Run: pip install psycopg2-binary")
sys.exit(1)
# Tables in dependency order (parents before children) to handle FK constraints.
# Tables without FK dependencies come first.
TABLES_ORDERED = [
"user",
"auth",
"config",
"document",
"migratehistory",
"alembic_version",
"memory",
"model",
"function",
"tool",
"skill",
"prompt",
"prompt_history",
"knowledge",
"file",
"knowledge_file",
"group",
"group_member",
"channel",
"channel_webhook",
"channel_member",
"message",
"message_reaction",
"channel_file",
"chat",
"chat_file",
"chat_message",
"chatidtag",
"tag",
"folder",
"feedback",
"note",
"api_key",
"oauth_session",
"access_grant",
]
# Columns that need INTEGER (0/1) -> BOOLEAN conversion
BOOL_COLUMNS = {
"auth": ["active"],
"function": ["is_active", "is_global"],
"model": ["is_active"],
"chat": ["archived", "pinned"],
"channel": ["is_private"],
"channel_member": ["is_active", "is_channel_muted", "is_channel_pinned"],
"chat_message": ["done"],
"folder": ["is_expanded"],
"message": ["is_pinned"],
"prompt": ["is_active"],
"skill": ["is_active"],
}
# config table: created_at/updated_at are DATETIME strings in SQLite,
# timestamp without time zone in PG — pass as-is, PG will parse them.
def get_sqlite_columns(sqlite_cur, table_name):
"""Get column names for a SQLite table."""
sqlite_cur.execute(f'PRAGMA table_info("{table_name}")')
return [row[1] for row in sqlite_cur.fetchall()]
def get_pg_columns(pg_cur, table_name):
"""Get column names for a PG table."""
pg_cur.execute(
"""
SELECT column_name FROM information_schema.columns
WHERE table_schema = 'public' AND table_name = %s
ORDER BY ordinal_position
""",
(table_name,),
)
return [row[0] for row in pg_cur.fetchall()]
def convert_row(table_name, columns, row):
"""Convert a SQLite row for PG insertion, handling type differences."""
converted = list(row)
bool_cols = BOOL_COLUMNS.get(table_name, [])
for i, col in enumerate(columns):
val = converted[i]
# INTEGER -> BOOLEAN conversion
if col in bool_cols and val is not None:
converted[i] = bool(val)
# Ensure JSON columns that are strings get passed as-is
# (psycopg2 handles this correctly with Json wrapper if needed,
# but PG json type accepts text strings directly)
return tuple(converted)
def migrate_table(sqlite_cur, pg_cur, table_name, batch_size=1000):
"""Migrate a single table from SQLite to PostgreSQL."""
# Get columns present in both databases
sqlite_cols = get_sqlite_columns(sqlite_cur, table_name)
pg_cols = get_pg_columns(pg_cur, table_name)
# Use intersection, preserving PG column order
common_cols = [c for c in pg_cols if c in sqlite_cols]
if not common_cols:
print(f" SKIP {table_name}: no common columns found")
return 0
# Extra columns in either side (for info)
sqlite_only = set(sqlite_cols) - set(pg_cols)
pg_only = set(pg_cols) - set(sqlite_cols)
if sqlite_only:
print(f" INFO: columns only in SQLite (will be skipped): {sqlite_only}")
if pg_only:
print(f" INFO: columns only in PG (will use defaults): {pg_only}")
# Read from SQLite
col_list_sqlite = ", ".join(f'"{c}"' for c in common_cols)
sqlite_cur.execute(f'SELECT {col_list_sqlite} FROM "{table_name}"')
total = 0
while True:
rows = sqlite_cur.fetchmany(batch_size)
if not rows:
break
converted_rows = [convert_row(table_name, common_cols, row) for row in rows]
# Build INSERT statement
col_list_pg = ", ".join(f'"{c}"' for c in common_cols)
placeholders = ", ".join(["%s"] * len(common_cols))
insert_sql = f'INSERT INTO "{table_name}" ({col_list_pg}) VALUES ({placeholders})'
psycopg2.extras.execute_batch(pg_cur, insert_sql, converted_rows, page_size=batch_size)
total += len(rows)
return total
def reset_sequences(pg_cur):
"""Reset auto-increment sequences for tables with serial/sequence PKs."""
sequences = [
("config", "config_id_seq", "id"),
("document", "document_id_seq", "id"),
("migratehistory", "migratehistory_id_seq", "id"),
]
for table, seq, col in sequences:
pg_cur.execute(f'SELECT COALESCE(MAX("{col}"), 0) FROM "{table}"')
max_val = pg_cur.fetchone()[0]
if max_val:
pg_cur.execute(f"SELECT setval('{seq}', {max_val})")
print(f" Sequence {seq} set to {max_val}")
def main():
parser = argparse.ArgumentParser(description="Migrate Open WebUI from SQLite to PostgreSQL")
parser.add_argument("--sqlite", required=True, help="Path to SQLite database file")
parser.add_argument("--pg", required=True, help="PostgreSQL connection string")
parser.add_argument("--batch-size", type=int, default=500, help="Batch size for inserts (default: 500)")
parser.add_argument("--dry-run", action="store_true", help="Show what would be done without writing")
args = parser.parse_args()
print(f"=== Open WebUI SQLite -> PostgreSQL Migration ===")
print(f"Source: {args.sqlite}")
print(f"Target: {args.pg.split('@')[0].split('://')[0]}://***@{args.pg.split('@')[-1]}")
print()
# Connect to SQLite
sqlite_conn = sqlite3.connect(args.sqlite)
sqlite_conn.row_factory = None # return tuples
sqlite_cur = sqlite_conn.cursor()
# Connect to PostgreSQL
pg_conn = psycopg2.connect(args.pg)
pg_cur = pg_conn.cursor()
# Verify both databases have the same alembic version
sqlite_cur.execute("SELECT version_num FROM alembic_version")
sqlite_ver = sqlite_cur.fetchone()
pg_cur.execute("SELECT version_num FROM alembic_version")
pg_ver = pg_cur.fetchone()
if sqlite_ver and pg_ver and sqlite_ver[0] != pg_ver[0]:
print(f"WARNING: Alembic version mismatch!")
print(f" SQLite: {sqlite_ver[0]}")
print(f" PostgreSQL: {pg_ver[0]}")
resp = input("Continue anyway? (y/N): ")
if resp.lower() != "y":
print("Aborted.")
sys.exit(1)
elif sqlite_ver and pg_ver:
print(f"Alembic version match: {sqlite_ver[0]}")
print()
if args.dry_run:
print("=== DRY RUN MODE ===")
print()
# Disable FK checks and truncate all target tables (reverse order)
print("Preparing target database...")
pg_cur.execute("SET session_replication_role = 'replica';") # disable FK triggers
for table in reversed(TABLES_ORDERED):
try:
if not args.dry_run:
pg_cur.execute(f'TRUNCATE TABLE "{table}" CASCADE')
print(f" TRUNCATE {table}")
except Exception as e:
pg_conn.rollback()
pg_cur.execute("SET session_replication_role = 'replica';")
print(f" TRUNCATE {table} - skipped ({e})")
if not args.dry_run:
pg_conn.commit()
print()
# Migrate each table
start_time = time.time()
total_rows = 0
errors = []
for table in TABLES_ORDERED:
try:
sqlite_cur.execute(f'SELECT count(*) FROM "{table}"')
count = sqlite_cur.fetchone()[0]
if count == 0:
print(f"[{table}] 0 rows - skip")
continue
print(f"[{table}] {count} rows...", end=" ", flush=True)
if args.dry_run:
print("(dry run)")
continue
migrated = migrate_table(sqlite_cur, pg_cur, table, args.batch_size)
pg_conn.commit()
print(f"OK ({migrated} migrated)")
total_rows += migrated
except Exception as e:
pg_conn.rollback()
# Re-disable FK triggers after rollback
pg_cur.execute("SET session_replication_role = 'replica';")
print(f"ERROR: {e}")
errors.append((table, str(e)))
print()
# Reset sequences
if not args.dry_run:
print("Resetting sequences...")
try:
reset_sequences(pg_cur)
pg_conn.commit()
except Exception as e:
pg_conn.rollback()
print(f" Sequence reset error: {e}")
# Re-enable FK triggers
pg_cur.execute("SET session_replication_role = 'origin';")
pg_conn.commit()
elapsed = time.time() - start_time
print()
print(f"=== Migration Complete ===")
print(f"Total rows migrated: {total_rows}")
print(f"Time: {elapsed:.1f}s")
if errors:
print(f"\nErrors ({len(errors)}):")
for table, err in errors:
print(f" {table}: {err}")
sys.exit(1)
# Verification
print("\n=== Verification ===")
for table in TABLES_ORDERED:
sqlite_cur.execute(f'SELECT count(*) FROM "{table}"')
s_count = sqlite_cur.fetchone()[0]
if s_count == 0:
continue
pg_cur.execute(f'SELECT count(*) FROM "{table}"')
p_count = pg_cur.fetchone()[0]
status = "OK" if s_count == p_count else f"MISMATCH (sqlite={s_count}, pg={p_count})"
print(f" {table}: {p_count} {status}")
sqlite_conn.close()
pg_cur.close()
pg_conn.close()
print("\nDone.")
if __name__ == "__main__":
main()
|
