Election errors

EventStoreDB
1

Hi,

The EventStore error log is currently filling up with this message:
[PID:05526:038 2016.01.04 00:34:40.437 ERROR JsonCodec ]
'POST /elections/viewchangeproof HTTP/1.1
Content-Type: application/json
Content-Length: 162
Expect: 100-continue
Host: 172.31.36.140:2112

{
  "serverId": "' is not a valid serialized
EventStore.Core.Messages.ElectionMessageDto+ViewChangeProofDto
Newtonsoft.Json.JsonReaderException: Unexpected character encountered
while parsing value: P. Path '', line 0, position 0.
  at Newtonsoft.Json.JsonTextReader.ParseValue () [0x00000] in
<filename unknown>:0
  at Newtonsoft.Json.JsonTextReader.ReadInternal () [0x00000] in
<filename unknown>:0
  at Newtonsoft.Json.JsonTextReader.Read () [0x00000] in <filename unknown>:0
  at Newtonsoft.Json.Serialization.JsonSerializerInternalReader.ReadForType
(Newtonsoft.Json.JsonReader reader,
Newtonsoft.Json.Serialization.JsonContract contract, Boolean
hasConverter) [0x00000] in <filename unknown>:0
  at Newtonsoft.Json.Serialization.JsonSerializerInternalReader.Deserialize
(Newtonsoft.Json.JsonReader reader, System.Type objectType, Boolean
checkAdditionalContent) [0x00000] in <filename unknown>:0

Is there anything I can do to fix it, or is it a bug in ES?

regards, Rickard

2

Sounds like your config... Putting it up would help.

3

Here's what I'm currently using, which includes my attempts at fixing
the timeouts mentioned in the other thread:

4

Why do you have external on ExtIp: 127.0.0.1

if you pull /gossip from a node :2112 what so you get?

5

This:
{
  "members": [
    {
      "instanceId": "99a2ec66-976f-4aad-a11e-48be83e5cf92",
      "timeStamp": "2016-01-04T10:03:42.842551Z",
      "state": "Master",
      "isAlive": true,
      "internalTcpIp": "172.31.36.140",
      "internalTcpPort": 1112,
      "internalSecureTcpPort": 0,
      "externalTcpIp": "127.0.0.1",
      "externalTcpPort": 1113,
      "externalSecureTcpPort": 0,
      "internalHttpIp": "172.31.36.140",
      "internalHttpPort": 2112,
      "externalHttpIp": "127.0.0.1",
      "externalHttpPort": 2113,
      "lastCommitPosition": 488961963,
      "writerCheckpoint": 488977390,
      "chaserCheckpoint": 488977390,
      "epochPosition": 481239063,
      "epochNumber": 121,
      "epochId": "f6f5fa05-a14a-4314-851e-dd0174cb2938",
      "nodePriority": 0
    },
    {
      "instanceId": "4e93ae9f-21cc-451a-84ff-6b3bb1360f9a",
      "timeStamp": "2016-01-04T10:03:41.405822Z",
      "state": "Slave",
      "isAlive": true,
      "internalTcpIp": "172.31.33.229",
      "internalTcpPort": 1112,
      "internalSecureTcpPort": 0,
      "externalTcpIp": "127.0.0.1",
      "externalTcpPort": 1113,
      "externalSecureTcpPort": 0,
      "internalHttpIp": "172.31.33.229",
      "internalHttpPort": 2112,
      "externalHttpIp": "127.0.0.1",
      "externalHttpPort": 2113,
      "lastCommitPosition": 488961963,
      "writerCheckpoint": 488977390,
      "chaserCheckpoint": 488977390,
      "epochPosition": 481239063,
      "epochNumber": 121,
      "epochId": "f6f5fa05-a14a-4314-851e-dd0174cb2938",
      "nodePriority": 0
    },
    {
      "instanceId": "72cf4b39-ac5d-42dc-8e8b-3eb7e9955ec4",
      "timeStamp": "2016-01-04T10:03:25.454771Z",
      "state": "Slave",
      "isAlive": true,
      "internalTcpIp": "172.31.33.228",
      "internalTcpPort": 1112,
      "internalSecureTcpPort": 0,
      "externalTcpIp": "127.0.0.1",
      "externalTcpPort": 1113,
      "externalSecureTcpPort": 0,
      "internalHttpIp": "172.31.33.228",
      "internalHttpPort": 2112,
      "externalHttpIp": "127.0.0.1",
      "externalHttpPort": 2113,
      "lastCommitPosition": 488961963,
      "writerCheckpoint": 488977390,
      "chaserCheckpoint": 488977390,
      "epochPosition": 481239063,
      "epochNumber": 121,
      "epochId": "f6f5fa05-a14a-4314-851e-dd0174cb2938",
      "nodePriority": 0
    }
  ],
  "serverIp": "172.31.33.229",
  "serverPort": 2112
}

Anything looks wrong there?

cheers, Rickard

6

Your gossip looks fine to me.

It would appear that your gossip messages are being cut off. I have seen this in the past. In the place where I have seen this occurring, the frequency was extremely low, like a couple of months apart.
What frequency are you seeing these messages in the logs?