|RockMongo and App||Marcin Oleszkiewicz||10/4/12 4:50 AM|
Without making any changes in main project classes and any mongo structures etc. from about a month I keep getting such errors:
Notice: Mongo::__construct() [<a href='mongo.--construct'>mongo.--construct</a>]: 127.0.0.1:27017: pool get (0x89dacf8) in C:\wamp\www\rockmongo\app\models\MServer.php on line 230
It is not only the problem of my application written in PHP because I also get those errors in RockMongo, I'm getting this error without any special conditions of app usage they occure one after another after restarting mongo and wamp, they offen kill wamp http, it happens on my local machine and on servers
I'm using on Windows 64b with WAMP
|Re: [mongodb-user] RockMongo and App||Hannes Magnusson||10/4/12 5:52 PM|
On Thu, Oct 4, 2012 at 4:50 AM, Marcin OleszkiewiczYou have MongoLog enabled with "fine" log level on the connection pool.
These noticed are mostly useful when you need to debug what is
happening internally in the driver when things don't work properly.
When enabled, these messages are very normal.
Search your codebase for "MongoLog::setLevel(...)" and
"MongoLog::setModule(..)" (case-in-senstive) and remove those lines if
you don't want to see the messages.
Alternatively, you can simply configure PHP to log to files, rather
then displaying them on the screen, if you still want to see these
messages for whatever reason.
|Re: RockMongo and App||Marcin Oleszkiewicz||10/4/12 10:43 PM|
And this is something normal that they kill Mongo databse and the WAMP server?
|Re: [mongodb-user] Re: RockMongo and App||Hannes Magnusson||10/5/12 9:26 AM|
No. Are you sure that is causing it?
And what exactly do you mean by "they kill Mongo database and the WAMP"?
> You received this message because you are subscribed to the Google
> Groups "mongodb-user" group.
> To post to this group, send email to mongod...@googlegroups.com
> To unsubscribe from this group, send email to
> See also the IRC channel -- freenode.net#mongodb
|Re: [mongodb-user] Re: RockMongo and App||Marcin Oleszkiewicz||10/6/12 5:02 AM|
Mongo stops responding and WAMP http services are going down and Apache is offline. I'm nearly sure that the reason is different. I've tripple checked and I'm not using any of MongoLog options inside of app. On production serv the 500 and 503's are caused by fucked up structure (other app on port 80 on IIS server and my app on 8081 apache via proxy by IIS) they are biting each other IIS doesn't like guest... but still I'm getting those notices from mongo pool also on my local machine where I have RockMongo and they occure inside the RockMongo
|Re: [mongodb-user] Re: RockMongo and App||Hannes Magnusson||10/6/12 1:23 PM|
Which version of the Mongo driver are you using exactly?
Have you tried turning the display_errors php.ini setting to off and
see if the services still fail?
|Re: RockMongo and App||Marcin Oleszkiewicz||10/7/12 11:37 PM|
This would be easier to debug if it would occure regularly in defined situations but it happens randomly... even if I will turn reporting to off I can't be sure if it is done or not... I thought that it's maybe more spread error and there are some reasonable reasons that can be found
RockMongo shows that my driver version is
and RockMongo itself is
|Re: [mongodb-user] Re: RockMongo and App||Hannes Magnusson||10/8/12 12:15 AM|
I checked out the v1.2.10 release.. This log message will always be
printed when a connection is created/retrieved from a pool.
This is fully and completely intentional log message and will only be
printed if you explicitly enable it by using MongoLog::set...() (along
with tons of other messages).
Are you sure you do not have any debug code automatically included
which enables it?
Any auto_prepend_file directives configured in PHP?
I'd recommend you upgrade to pecl/mongo version 1.2.12 anyway, but it
shouldn't change anything for this specific case.