Apache flume does not read logs line by line when pushed from heka using multiport syslog TCP as source - flume

Question

I am using heka to push the logs to flume.
In heka I am able to see the logs getting tagged event by event(line by line)
But in flume I am not getting line by line. It merges 2 or 3 lines of log together and prints it as a chunk of message.
I am using source as multiport_syslogtcp and channel as memory channel, sink as file_roll.
Flume conf file:
a1.sources = r1
a1.sinks = k1
a1.channels = c1
a1.sources.r1.type = multiport_syslogtcp
a1.sources.r1.host = 0.0.0.0
a1.sources.r1.ports = 44444
a1.sinks.k1.type = file_roll
a1.sinks.k1.channel = c1
a1.sinks.k1.sink.directory = /var/log/flumelog
a1.sinks.k1.sink.rollInterval = 0
a1.sources.r1.channels = c1
a1.sinks.k1.channel = c1
Example log line:
{ "name": "SIDXDropped", "host": "astr", "date": "2016-08-10T10:46:13.419Z", "schema": "mce-data", "source": "MOS_MCE_VM:worker", "type": "Application", "subtype": "", "context": "", "category": "LIVE.capture", "severity": "Critical", "description": "", "location": "node:2561", "data": {"event":{"eventData":{"channelName":"ahdenclinearwf/53560_abr","eventDetail":"RepID: 2000000_audio_frag_302 SIDX: {reference_ID: 302; PTS: 323565332463; first_offset: 72970420204; size: 821936; duration: 194400; timescale: 90000","eventName":"SIDXDropped","level":"MAJOR","profileName":"1864k","serviceType":"LIVE","subType":"CIFConverter","type":"RECORDER"}}}, "alarmop": "" }
{ "name": "SIDXDropped", "host": "astr", "date": "2016-08-10T10:46:13.419Z", "schema": "mce-data", "source": "MOS_MCE_VM:worker", "type": "Application", "subtype": "", "context": "", "category": "LIVE.capture", "severity": "Critical", "description": "", "location": "node:2561", "data": {"event":{"eventData":{"channelName":"ahdenclinearwf/53560_abr","eventDetail":"RepID: 2000000_audio_frag_302 SIDX: {reference_ID: 302; PTS: 323565332463; first_offset: 72970420204; size: 821936; duration: 194400; timescale: 90000","eventName":"SIDXDropped","level":"MAJOR","profileName":"1864k","serviceType":"LIVE","subType":"CIFConverter","type":"RECORDER"}}}, "alarmop": "" }
Example output in heka:
2016/08/26 09:00:58 {"message":"{ \"name\": \"SIDXDropped\", \"host\": \"astr\", \"date\": \"2016-08-10T10:46:13.419Z\", \"schema\": \"mce-data\", \"source\": \"MOS_MCE_VM:worker\", \"type\": \"Application\", \"subtype\": \"\", \"context\": \"\", \"category\": \"LIVE.capture\", \"severity\": \"Critical\", \"description\": \"\", \"location\": \"node:2561\", \"data\": {\"event\":{\"eventData\":{\"channelName\":\"ahdenclinearwf\/53560_abr\",\"eventDetail\":\"RepID: 2000000_audio_frag_302 SIDX: {reference_ID: 302; PTS: 323565332463; first_offset: 72970420204; size: 821936; duration: 194400; timescale: 90000\",\"eventName\":\"SIDXDropped\",\"level\":\"MAJOR\",\"profileName\":\"1864k\",\"serviceType\":\"LIVE\",\"subType\":\"CIFConverter\",\"type\":\"RECORDER\"}}}, \"alarmop\": \"\" }","#timestamp":"2016-08-26T09:00:58.494Z","subsystem":"cls","host":"cls-1","severity":7,"component":"mce","logger":"mos-mce","logsource":"event"}
2016/08/26 09:00:58 {"message":"{ \"name\": \"SIDXDropped\", \"host\": \"astr\", \"date\": \"2016-08-10T10:46:13.419Z\", \"schema\": \"mce-data\", \"source\": \"MOS_MCE_VM:worker\", \"type\": \"Application\", \"subtype\": \"\", \"context\": \"\", \"category\": \"LIVE.capture\", \"severity\": \"Critical\", \"description\": \"\", \"location\": \"node:2561\", \"data\": {\"event\":{\"eventData\":{\"channelName\":\"ahdenclinearwf\/53560_abr\",\"eventDetail\":\"RepID: 2000000_audio_frag_302 SIDX: {reference_ID: 302; PTS: 323565332463; first_offset: 72970420204; size: 821936; duration: 194400; timescale: 90000\",\"eventName\":\"SIDXDropped\",\"level\":\"MAJOR\",\"profileName\":\"1864k\",\"serviceType\":\"LIVE\",\"subType\":\"CIFConverter\",\"type\":\"RECORDER\"}}}, \"alarmop\": \"\" }","#timestamp":"2016-08-26T09:00:58.494Z","subsystem":"cls","host":"cls-1","severity":7,"component":"mce","logger":"mos-mce","logsource":"event"}
Equivalent messages in file in flume:
{"message":"{ \"name\": \"SIDXDropped\", \"host\": \"astr\", \"date\": \"2016-08-10T10:46:13.419Z\", \"schema\": \"mce-data\", \"source\": \"MOS_MCE_VM\", \"type\": \"Application\", \"subtype\": \"\", \"context\": \"\", \"category\": \"LIVE.capture\", \"severity\": \"Critical\", \"description\": \"\", \"location\": \"node:2561\", \"data\": {\"event\":{\"eventData\":{\"channelName\":\"ahdenclinearwf\/53560_abr\",\"eventDetail\":\"RepID: 2000000_audio_frag_302 SIDX: {reference_ID: 302; PTS: 323565332463; first_offset: 72970420204; size: 821936; duration: 194400; timescale: 90000\",\"eventName\":\"SIDXDropped\",\"level\":\"MAJOR\",\"profileName\":\"1864k\",\"serviceType\":\"LIVE\",\"subType\":\"CIFConverter\",\"type\":\"RECORDER\"}}}, \"alarmop\": \"\" }","#timestamp":"2016-08-26T06:58:06.743Z","subsystem":"cls","host":"cls-1","severity":7,"component":"mce","logger":"mos-mce","logsource":"event"}{"message":"{ \"name\": \"SIDXDropped\", \"host\": \"astr\", \"date\": \"2016-08-10T10:46:13.419Z\", \"schema\": \"mce-data\", \"source\": \"MOS_MCE_VM\", \"type\": \"Application\", \"subtype\": \"\", \"context\": \"\", \"category\": \"LIVE.capture\", \"severity\": \"Critical\", \"description\": \"\", \"location\": \"node:2561\", \"data\": {\"event\":{\"eventData\":{\"channelName\":\"ahdenclinearwf\/53560_abr\",\"eventDetail\":\"RepID: 2000000_audio_frag_302 SIDX: {reference_ID: 302; PTS: 323565332463; first_offset: 72970420204; size: 821936; duration: 194400; timescale: 90000\",\"eventName\":\"SIDXDropped\",\"level\":\"MAJOR\",\"profileName\":\"1864k\",\"serviceType\":\"LIVE\",\"subType\":\"CIFConverter\",\"type\":\"RECORDER\"}}}, \"alarmop\": \"\" }","#timestamp":"2016-08-26T09:00:58.494Z","subsystem":"cls","host":"cls-1","severity":7,"component":"mce","logger":"mos-mce","logsource":"event"}
The output log in flume does not print logs line by line as received by heka.
Could you please suggest if anything has to be changed to in flume source or memory channel