User and support list for MongoDB, a NoSQL database. ()

Eliot Horowitz | 1 Dec 2011 01:16

Re: Re: slow db connect

Looks like someone called shutdown on it:
Tue Nov 29 12:18:42 [conn4300] terminating, shutdown command received

But it failed to shutdown cleanly.

If you have more of the log might be able to see where the shutdown came from.

On Wed, Nov 30, 2011 at 7:10 PM, Mason <rhymejerky@...> wrote:
> Turned out that 1 of the mongos was broken.  When checking the logs,
> the broken mongos has the following in log
>
> Tue Nov 29 12:18:24 [Balancer] distributed lock 'balancer/
> ii72-14.friendfinderinc.com:57017:1321512829:1804289383' acquired,
> ts : 4ed53e0fc3b16b66a82f3051
> Tue Nov 29 12:18:24 [Balancer] distributed lock 'balancer/
> ii72-14.friendfinderinc.com:57017:1321512829:1804289383' unlocked.
> Tue Nov 29 12:18:34 [Balancer] distributed lock 'balancer/
> ii72-14.friendfinderinc.com:57017:1321512829:1804289383' acquired,
> ts : 4ed53e1ac3b16b66a82f3052
> Tue Nov 29 12:18:34 [Balancer] distributed lock 'balancer/
> ii72-14.friendfinderinc.com:57017:1321512829:1804289383' unlocked.
> Tue Nov 29 12:18:42 [conn4300] terminating, shutdown command received
> Tue Nov 29 12:18:42 [conn4300] dbexit: shutdown called rc:0 shutdown
> called
> Logstream::get called in uninitialized state
> Tue Nov 29 12:18:42 [mongosMain] dbexit:  rc:0
> *** glibc detected *** /usr/bin/mongos: free(): invalid pointer:
> 0x0000000000b0d380 ***
> ======= Backtrace: =========
> /lib64/libc.so.6[0x3980a71684]

(Continue reading)

headers

Mason | 1 Dec 2011 04:07

Re: slow db connect

I see these in the log, so that must be the shut down connection.
Does mongosMain mean the shell?  I don't have anything running on
localhost at port 57443

Tue Nov 29 12:17:21 [mongosMain] connection accepted from
127.0.0.1:57443 #4300
Tue Nov 29 12:18:42 [conn4300] terminating, shutdown command received
Tue Nov 29 12:18:42 [conn4300] dbexit: shutdown called rc:0 shutdown
called

I was able to do "mongo -port <port for mongos" on the 'half' shutdown
instance, but never got the shell prompt.  If that mongos is still
part of the my connect string, would it still try to connect to it
first before trying another mongos instance?  I consistently see it
took 30 secs to connect, is this a set timeout value before trying
another mongos?  Also, does the PERL driver allow persistent
connection?

headers

Richard Kreuter | 1 Dec 2011 23:37

Re: slow db connect

Was the broken mongos still running, or had the process gone away when
you were trying to connect to it?

On Nov 30, 10:07 pm, Mason <rhymeje...@...> wrote:
> I see these in the log, so that must be the shut down connection.
> Does mongosMain mean the shell?  I don't have anything running on
> localhost at port 57443
>
> Tue Nov 29 12:17:21 [mongosMain] connection accepted from
> 127.0.0.1:57443 #4300
> Tue Nov 29 12:18:42 [conn4300] terminating, shutdown command received
> Tue Nov 29 12:18:42 [conn4300] dbexit: shutdown called rc:0 shutdown
> called
>
> I was able to do "mongo -port <port for mongos" on the 'half' shutdown
> instance, but never got the shell prompt.  If that mongos is still
> part of the my connect string, would it still try to connect to it
> first before trying another mongos instance?  I consistently see it
> took 30 secs to connect, is this a set timeout value before trying
> another mongos?  Also, does the PERL driver allow persistent
> connection?

--

-- 
You received this message because you are subscribed to the Google Groups "mongodb-user" group.
To post to this group, send email to mongodb-user@...
To unsubscribe from this group, send email to mongodb-user+unsubscribe <at> googlegroups.com.
For more options, visit this group at http://groups.google.com/group/mongodb-user?hl=en.