-
Type: Bug
-
Resolution: Cannot Reproduce
-
Priority: Minor - P4
-
Affects Version/s: None
-
Component/s: None
-
Labels:
-
Empty show more show less
We have a small cluster of two instances with an aditional arbiter.
There are multiple node clients that are connected to the cluster.
On friday morning one client lost the connection to one of the instances: ("MongoError: server xxx received an error
{'name':'MongoError','message':'read ECONNRESET'}")
It apparently managed to reconnect as for a few hours everything went fine, but than the same client only failed constantly on every query with the error message: "MongoError: not master and slaveOk=false"
Why hasnt the client reconnected to the master?
And why did this issue came up delayed and not at the same time as the resetted connection?
Thanks in advance!