6.1.2 to 6.2.1 failed to upgrade SQLlite DB for me :’(
[seafile@birdie upgrade]$ sh upgrade_6.1_6.2.sh
This script would upgrade your seafile server from 6.1 to 6.2
Press [ENTER] to contiune
Updating seafile/seahub database …
[INFO] You are using SQLite3
[INFO] updating seahub database…
Traceback (most recent call last):
File “/home/sites/teufou.net/seafile/seafile-server-6.2.1/upgrade/db_update_helper.py”, line 363, in
main()
File “/home/sites/teufou.net/seafile/seafile-server-6.2.1/upgrade/db_update_helper.py”, line 358, in main
db_updater.update_db()
File “/home/sites/teufou.net/seafile/seafile-server-6.2.1/upgrade/db_update_helper.py”, line 259, in update_db
super(SQLiteDBUpdater, self).update_db()
File “/home/sites/teufou.net/seafile/seafile-server-6.2.1/upgrade/db_update_helper.py”, line 120, in update_db
self.update_seahub_sql(seahub_sql)
File “/home/sites/teufou.net/seafile/seafile-server-6.2.1/upgrade/db_update_helper.py”, line 283, in update_seahub_sql
self.apply_sqls(self.seahub_db, sql_path)
File “/home/sites/teufou.net/seafile/seafile-server-6.2.1/upgrade/db_update_helper.py”, line 273, in apply_sqls
conn.execute(line)
sqlite3.OperationalError: no such table: sysadmin_extra_userloginlog
Just a little question: Testing WSGI mode here (with 6.1.2 version, 6.2.1 isnt out for rpi), but thats working fine. The only problem is login page, after typing user and password it didnt redirect to library page. But login works, as if i manually type server adress (which is https://xyz:8001) i get library page. Nginx config is same like posted above, nothing else changed, with fcgi login works normal.
Maybe thats only while using older 6.1.2 version and i have to wait for new 6.2.1? or did i something wrong?
I’m having the same problem as you and the others are describing. Can you elaborate on this step? Thanks. I’m not exactly sure on how to proceed to carry out what you wrote.
6.2.1 was released as stable before the 14 days of beta time were over. 6.2.1 still has some bugs, hence is still rather beta than stable.
If you state on your website that you run 2 weeks in beta please stick to it. It is not consequent and gives an impression that you don’t care much about what you tell/sell your clients.
I don’t know why table sysadmin_extra_userloginlog is missing in your sqlite3 db. Anyway, you can create it by yourself via the following sentence:
CREATE TABLE "sysadmin_extra_userloginlog" ("id" integer NOT NULL PRIMARY KEY AUTOINCREMENT, "username" varchar(255) NOT NULL, "login_date" datetime NOT NULL, "login_ip" varchar(128) NOT NULL);
CREATE INDEX "sysadmin_extra_userloginlog_14c4b06b" ON "sysadmin_extra_userloginlog" ("username");
CREATE INDEX "sysadmin_extra_userloginlog_28ed1ef0" ON "sysadmin_extra_userloginlog" ("login_date");
@Captain_Rage I don’t know and have not checked if 6.2.2 fixed this already, but what I had to run was sqlite3 seahub.db < seafile-server-latest/upgrade/sql/6.2.0/sqlite/seahub.sql.
It is for “about 2 weeks”, not for exactly 2 weeks. As the SQLite upgrade script bug is urgent, we publish a new version sooner than waiting 2 weeks. As I don’t see there are other significant bugs, I remove the beta mark.
Not fixed in 6.2.2
Updating seafile/seahub database …
[INFO] You are using SQLite3
[INFO] updating seahub database…
Traceback (most recent call last):
File “/home/sites/teufou.net/seafile/seafile-server-6.2.2/upgrade/db_update_helper.py”, line 363, in
main()
File “/home/sites/teufou.net/seafile/seafile-server-6.2.2/upgrade/db_update_helper.py”, line 358, in main
db_updater.update_db()
File “/home/sites/teufou.net/seafile/seafile-server-6.2.2/upgrade/db_update_helper.py”, line 259, in update_db
super(SQLiteDBUpdater, self).update_db()
File “/home/sites/teufou.net/seafile/seafile-server-6.2.2/upgrade/db_update_helper.py”, line 120, in update_db
self.update_seahub_sql(seahub_sql)
File “/home/sites/teufou.net/seafile/seafile-server-6.2.2/upgrade/db_update_helper.py”, line 283, in update_seahub_sql
self.apply_sqls(self.seahub_db, sql_path)
File “/home/sites/teufou.net/seafile/seafile-server-6.2.2/upgrade/db_update_helper.py”, line 273, in apply_sqls
conn.execute(line)
sqlite3.OperationalError: no such table: sysadmin_extra_userloginlog
I have migrated to 6.2 successfully but noticed when using WSGI mode that seahub is exposed on port 8000 on 0.0.0.0 not 127.0.0.1. As I’m on a VPS with no iptables this is a problem.
I have changed the 0.0.0.0 to 127.0.0.1 which works for me. Is there another way which is more suitable?
(BTW must try and learn to format code in these fancy editors on day)
EDIT: Just noticed user TMHBOFH has also spotted this.
Best Regards
For people not configured Nginx/Apache, ./seahub.sh start need to listen on 0.0.0.0 to work. For not broken existing Seafile installations, this should be default behaviour. We will add an option to listen on 127.0.0.1 only.
If your Seafile installation is behind Apache/Nginx and you implement WSGI then when Seahub is run it will open port 8000 on 0.0.0.0.
If, like me, you have no firewall then this exposes port 8000 to the 'net and it is not required to be so. So the answer is yes to the question of modifying seahub.sh if you consider this a risk. Also bear in mind that any changes to seahub.sh will disappear in the next update.
I adapted the rpi configuration on Github to consider proxy_http_ version 1.1 if we get such error"connection refused...HTTP/1.1." on nginx logfiles or we have http2. The other should be done by development team.