Commit Graph

12 Commits (5b756e0fbefe83145fed421887926a7ef62a45c9)

Author SHA1 Message Date
Jason Wilder 5abdb42a7d Use a bounded pool for remote writes
Under highly conncurrent write load, the coordinating node would
create a connection to any other node that is part of the replica
group.  Since each connection can be expensive, OOM sitations could
occur because there was no bounds on the number of new connections
that would be created.  If writes on a remote node were slow, connections
could pile up an exacerbate the problem.

This switches the pool to be bounded and has a checkout that is blocking
with a timeout.  If a connection is available, it's returned immediately.
If the pool still has room for more connections, it will create one if needed.
Otherwise, the call will block until a connection becomes available or
the timeout expires.  In the case of a timeout, it is propogated back up
to the PointsWriter that determine what do return to the client.
2016-01-26 17:08:36 -07:00
Jason Wilder f58f0f5373 Fix cluster tests 2016-01-21 15:28:34 -05:00
Jason Wilder 0926b19e6b Prevent creating points with NaN float values
Float values are not supported in the existing engine and the tsm1
engines.  This changes NewPoint to return an error if a field value
contains a NaN field.  It also allows us to validate fields to prevent
other unsupported types from sneaking in through other input plugins.
2015-10-27 17:12:52 -06:00
Cory LaNou d19a510ad2 refactor Points and Rows to dedicated packages 2015-09-16 15:33:08 -05:00
Philip O'Toole a19cea36bd Rename cluster unit test function
Makes future tests, related to shard mapping, clearer.
2015-07-17 13:05:15 -07:00
Jason Wilder eb1cd44b8d Log write errors
Since the client only receives a "write failed" or "partial write" error
message, log more context in the logs.
2015-06-09 14:49:22 -06:00
Jason Wilder 1024965db7 Create shard received from cluster writer 2015-06-05 22:16:51 -06:00
Ben Johnson fb06549552 remove bind address from cluster config 2015-06-05 17:07:54 -06:00
Ben Johnson 5a5c077790 refactor cluster to use mux 2015-06-05 16:54:12 -06:00
Jason Wilder 3957e096f8 Remove ownerID from protobufs
Not needed since the node that processes the request is the owner.
2015-06-02 14:45:52 -06:00
Ben Johnson bf823d9887 Integrating cmd/influxd/run. 2015-05-30 14:06:36 -06:00
Ben Johnson c916256ac9 Rename cluster.Writer to cluster.ShardWriter. 2015-05-30 14:05:27 -06:00