DB->put: attempt to modify a read-only database" - in logs with every SET

58 views

Skip to first unread message

RaVbaker

unread,

Dec 19, 2009, 4:43:43 AM12/19/09

to memcachedb

Hello,
I'm a newbie in memcachedb. It was quite easy to set up memcachedb
server with master/slave replication on our machines.

It works quite good. But suddenly I start getting error "DB->put:
attempt to modify a read-only database" in memcachedb logs in every
SET. GETs stop responding, SETs also.

I'm using version 1.2.0 and it runs this way:
memcachedb -p 21201 -n 2 -R 0.0.0.0 31201 -M -m 64 -A 4096 -u
memcachedb -l 0.0.0.0 -H /var/lib/memcachedb -f /var/lib/memcachedb/
default.db

Maybe a few lines from logs can help diagnosis my problem:

[memcachedb] [Fri Dec 18 16:57:41 2009] "memp_trickle thread: writing
0 dirty pages"
[memcachedb] [Fri Dec 18 16:58:01 2009] "event:
DB_EVENT_REP_PERM_FAILED, insufficient acks, the master will flush the
txn log buffer"
[memcachedb] [Fri Dec 18 16:58:06 2009] "event: DB_EVENT_REP_CLIENT,
I<0.0.0.0:31201> am now a replication client."
[memcachedb] [Fri Dec 18 16:58:06 2009] "DB_ENV->rep_elect:WARNING:
nvotes (1) is sub-majority with nsites (2)"
[memcachedb] [Fri Dec 18 16:58:07 2009] "event:
DB_EVENT_REP_NEWMASTER, a new master<eid: 0> has been established, but
not me<0.0.0.0:31201>"
[memcachedb] [Fri Dec 18 16:58:11 2009] "memp_trickle thread: writing
0 dirty pages"
[memcachedb] [Fri Dec 18 16:58:12 2009] "event:
DB_EVENT_REP_NEWMASTER, a new master<eid: 0> has been established, but
not me<0.0.0.0:31201>"
[memcachedb] [Fri Dec 18 16:59:13 2009] "DB_ENV handle waiting 1
minutes for replication lockout to complete"
[memcachedb] [Fri Dec 18 16:59:14 2009] "DB->put: attempt to modify a
read-only database"