Seafile Server broken after system upgrade?

server

#1

hey there!
i’ve upgraded my debian server and now seafile is unable to start correctly. so i’ve upgraded seafile as well (in hope it would fix the issue), but it’s still not working. any ideas?

env: Debian 9.3/Linux 4.9.0-5-amd64
seafile server version: 6.3.2

the web-part of the server works (i can see the site and even log in); but the clients are unable to see the libraries (even through the web-interface).

any help is appreciated! thanks for your time!

==> logs/ccnet.log <==
[01/12/18 14:35:59] ../common/peer.c(943): Local peer down
[01/12/18 14:35:59] ../common/peer.c(943): Local peer down
[01/12/18 14:36:09] ../common/session.c(398): Accepted a local client
[01/12/18 14:36:09] ../common/session.c(398): Accepted a local client
[01/12/18 14:36:09] ../common/peer.c(943): Local peer down
[01/12/18 14:36:09] ../common/peer.c(943): Local peer down
[01/12/18 14:36:19] ../common/session.c(398): Accepted a local client
[01/12/18 14:36:19] ../common/session.c(398): Accepted a local client
[01/12/18 14:36:19] ../common/peer.c(943): Local peer down
[01/12/18 14:36:19] ../common/peer.c(943): Local peer down

==> logs/controller.log <==
[01/12/18 14:36:09] seafile-controller.c(396): pid file /var/www/seafile/pids/seaf-server.pid does not exist
[01/12/18 14:36:09] seafile-controller.c(416): seaf-server need restart...
[01/12/18 14:36:09] seafile-controller.c(186): starting seaf-server ...
[01/12/18 14:36:09] seafile-controller.c(73): spawn_process: seaf-server -F /var/www/seafile/conf -c /var/www/seafile/ccnet -d /xyz/web/seafile/seafile-data -l /var/www/seafile/logs/seafile.log -P /var/www/seafile/pids/seaf-server.pid
[01/12/18 14:36:09] seafile-controller.c(88): spawned seaf-server, pid 19293
[01/12/18 14:36:19] seafile-controller.c(396): pid file /var/www/seafile/pids/seaf-server.pid does not exist
[01/12/18 14:36:19] seafile-controller.c(416): seaf-server need restart...
[01/12/18 14:36:19] seafile-controller.c(186): starting seaf-server ...
[01/12/18 14:36:19] seafile-controller.c(73): spawn_process: seaf-server -F /var/www/seafile/conf -c /var/www/seafile/ccnet -d /xyz/web/seafile/seafile-data -l /var/www/seafile/logs/seafile.log -P /var/www/seafile/pids/seaf-server.pid
[01/12/18 14:36:19] seafile-controller.c(88): spawned seaf-server, pid 19315

==> logs/seafile.init.log <==

 About to perform stop for seafile at 2018-01-12T14:32:20+01:00

-e

 About to perform start for seafile at 2018-01-12T14:32:22+01:00


Seafile controller is already running, pid 15949


==> logs/seafile.log <==
[01/12/2018 02:35:49 PM] http-server.c(1915): Could not bind socket: Address already in use
[01/12/2018 02:35:59 PM] ../common/mq-mgr.c(60): [mq client] mq cilent is started
[01/12/2018 02:35:59 PM] ../common/mq-mgr.c(106): [mq mgr] publish to hearbeat mq: seaf_server.heartbeat
[01/12/2018 02:35:59 PM] http-server.c(1915): Could not bind socket: Address already in use
[01/12/2018 02:36:09 PM] ../common/mq-mgr.c(60): [mq client] mq cilent is started
[01/12/2018 02:36:09 PM] ../common/mq-mgr.c(106): [mq mgr] publish to hearbeat mq: seaf_server.heartbeat
[01/12/2018 02:36:09 PM] http-server.c(1915): Could not bind socket: Address already in use
[01/12/2018 02:36:19 PM] ../common/mq-mgr.c(60): [mq client] mq cilent is started
[01/12/2018 02:36:19 PM] ../common/mq-mgr.c(106): [mq mgr] publish to hearbeat mq: seaf_server.heartbeat
[01/12/2018 02:36:19 PM] http-server.c(1915): Could not bind socket: Address already in use

==> logs/seahub_django_request.log <==
    quota_total = seafile_api.get_user_quota(email)
  File "/var/www/seafile/seafile-server-6.2.3/seafile/lib64/python2.7/site-packages/seaserv/api.py", line 614, in get_user_quota
    return seafserv_threaded_rpc.get_user_quota(username)
  File "/var/www/seafile/seafile-server-6.2.3/seafile/lib64/python2.7/site-packages/pysearpc/client.py", line 110, in newfunc
    ret_str = self.call_remote_func_sync(fcall_str)
  File "/var/www/seafile/seafile-server-6.2.3/seafile/lib64/python2.7/site-packages/ccnet/rpc.py", line 75, in call_remote_func_sync
    req_id = self._start_service(client)
  File "/var/www/seafile/seafile-server-6.2.3/seafile/lib64/python2.7/site-packages/ccnet/rpc.py", line 36, in _start_service
    raise SearpcError("Error received: %s %s (In _start_service)" % (rsp.code, rsp.code_msg))
SearpcError: Error received: 511 Unknown service (In _start_service)

==> logs/seahub.init.log <==
 About to perform start for seahub at 2018-01-12T14:32:22+01:00


LC_ALL is not set in ENV, set to en_US.UTF-8
Starting seahub (fastcgi) at 127.0.0.1:8081 ...

Seahub is started

Done.

#2

Maybe You should delete seafile cache.
rm -rf /tmp/seahub_cache
Or, just to make sure this does not crash anything, just move it elsewhere with mv command. And put it back if this crashes your Seafile installation.

Afterwards restart both services.


#3

Hi ruffni,

try to stop all seafile and seahub processes first. In necessary even with “kill”. Then try to start seafile and seahub.

If that does not help, please check the permissions of the seafile folders. Probably you performed the update as root and seafile has no more access to necessary files.

Best regards
Christoph


#4

Hi, I have the same problem (but I don’t know if the system upgrade is the cause in my case, as I did this already 6 days ago and Seafile worked until I upgraded it).
Yesterday, I upgraded Seafile from 5.1.8 to 6.2.7 and it worked fine for some hours. I could access all files and change settings via seahub. Then I rebooted my server, since then I get exactly the same error messages as you, “Local peer down” in ccnet.log, “could not bind socket” in seafile.log and those django errors (those seem to be a consequence of the could not bind socket error, see https://github.com/haiwen/ccnet/issues/107#issuecomment-213947818).

Have you been able to solve this yet?


#5

thanks for all the replies

i tried pretty much all of that.
@Henrik
i moved the cache to a new place, there’s a new one now, but seafile is still not working (correction; the web-interface works, and even allowed me to login, but doesn’t show any of the files).
@christophdb
i also tried restarting the services a couple of times, no luck whatsoever… and the logs remain only semi-helpful.


#6

@mli
actually, my guess is also an incompatibility with the upgrade to linux 4.9.0-5, cause seafile just worked til then. also tried to upgrade the installation (to 6.2.x), but no luck at all.
so i’m stuck as you. have you already tried installing the whole thing freshly? this would be my next step - and afterwards trying to migrate the data / change back the settings to what they are now.


#7

Have you tried to find out, what process is listening on ports needed by Seafile sockets?
E.g. by typing:
sudo netstat -nap | grep LISTEN