Constant problem with Couchbase 2.0.1

Hi all,

I have a serious issue on Couchbase cluster. I have 2 nodes on Linux Couchbase Enterprise 2.0.1 and each day I keep getting this error on random times (I think when there is high burst of gets/sets to the system):

This happens first:

Control connection to memcached on 'ns_1@XXXXXXXX' disconnected: {{badmatch,
{error,
closed}},
[{mc_client_binary,
cmd_binary_vocal_recv,
5},
{mc_client_binary,
select_bucket,
2},
{ns_memcached,
ensure_bucket,
2},
{ns_memcached,
handle_info,
2},
{gen_server,
handle_msg,
5},
{proc_lib,
init_p_do_apply,
3}]}

Thenk this:

Port server memcached on node 'ns_1@XXXXXXX' exited with status 134. Restarting. Messages: Wed Oct 16 08:32:33.104109 CEST 3: 410 Closing connection due to read error: Connection reset by peer

It then restarts one of the nodes. Thing is that when the node restarts it needs about 5000 seconds to start one of the buckets on server.

This is the bucket:

NODES: 2
ITEMS: 15289871
RAM QUOTA: 8.75GB /13.1GB
DATA/DISK USAGE: 40.7GB / 75.5GB

We have similar setup on version 1.8.1 and it works flawlessly with 6.000.000 items. We just don't understand what is happening here. Please any kind of help is much appreciated.

1 Answer

« Back to question.

Hello,

Any chance to move to the latest release? 2.2.x

To be able to help you, we will need more information could you please follow the steps documented here:
http://www.couchbase.com/wiki/display/couchbase/Working+with+the+Couchba...

and let us/me know where is the final log file.

Regards
Tug
@tgrall