-
Type:
Bug
-
Resolution: Incomplete
-
Priority:
Critical - P2
-
None
-
Affects Version/s: 2.0.3
-
Component/s: MapReduce
-
None
-
Environment:Linux 3.0.0-1-amd64 #1 SMP Sat Aug 27 16:21:11 UTC 2011 x86_64 GNU/Linux
-
Linux
-
None
-
0
-
None
-
None
-
None
-
None
-
None
-
None
Instead of
%database - real db name
%collection - real collection name (~ 26 millions of rows)
> db.currentOp().inprog.forEach(function(row) { if (row.active) print(tojson(row)); } );
{
"opid" : 16617560,
"active" : true,
"lockType" : "read",
"waitingForLock" : false,
"secs_running" : 7941,
"op" : "query",
"ns" : "%database.%collection",
"query" : {
"$msg" : "query not recording (too large)"
},
"client" : "127.0.0.1:45898",
"desc" : "conn",
"threadId" : "0x7f8aa72f2700",
"connectionId" : 84024,
"msg" : "m/r: (1/3) emit phase",
"numYields" : 19082
}
$ ps aux | grep mongod
mongodb 5400 6.2 44.1 151646464 14563872 ? Sl Mar20 197:08 /usr/bin/mongod --dbpath /work1/mongodb --logpath /var/log/mongodb-local/mongodb-local.log --config /etc/mongodb-local.conf run
> getMemInfo()
{ "virtual" : 113, "resident" : 76 }
> db.stats()
# .......... wait to long, and nothing
> db.currentOp().inprog.length
19998
$ tail /var/log/mongodb-local/mongodb-local.log
Thu Mar 22 13:22:50 [initandlisten] connection accepted from 127.0.0.1:51562 #633178
Thu Mar 22 13:22:50 [initandlisten] connection refused because too many open connections: 20000
Thu Mar 22 13:22:50 [initandlisten] connection accepted from 127.0.0.1:51563 #633179
Thu Mar 22 13:22:50 [initandlisten] connection refused because too many open connections: 20000
Thu Mar 22 13:22:50 [initandlisten] connection accepted from 127.0.0.1:51564 #633180
Thu Mar 22 13:22:50 [initandlisten] connection refused because too many open connections: 20000
Thu Mar 22 13:22:50 [initandlisten] connection accepted from 127.0.0.1:51565 #633181
Thu Mar 22 13:22:50 [initandlisten] connection refused because too many open connections: 20000
Thu Mar 22 13:22:50 [initandlisten] connection accepted from 127.0.0.1:51566 #633182
And now I kill this MR job
> db.killOp(16617560)
{ "info" : "attempting to kill op" }
> db.currentOp().inprog.length
19502
> db.currentOp().inprog.length
17396