question

Upvotes
Accepted
julien.dominici avatar image
112 ●2 ●5 ●6

Packet losses with RRCPDL under trep3.3.0 (linux rhel7) and local ATS setup

Hi dear developpers team and community, this if following case 07426946 where I was asked to post my question here.

I deployed a new ATS + ADH + ADS setup today in production, and I am continuously seeing this error message into the RRCPDL-Source logs :

----- Fri Mar 08 10:58:34.871198 2019 /tmp/.rrcp/source.0.rrmp: WARNING: [../Wrapper/Userlevel/rrcpCW_NetMgr.c,NetMgr_sendPkt(),274] error writing to the network: ------

Then ADS logs packet losses :

------- RRCP STATUS MSG: RRCP_BC_MISSEDMSGS: gap in broadcast msgs from node 169752968 (10.30.57.136) From Port 37000 --------

The RRCPDL parameters are :

------ [pars3pmdsc527:/rmds/users/mdadmin] rrdump source -DL -port 37000 -P

rrdump v6.7.F30: [423]:Connected with RRCP-daemonless, Version rrcp6.7.T25(6070025), Control Port: 37000

RRCP Parameter Values (IpAddr: 10.30.57.136 (169752968) source mode, device /tmp/.rrcp/source.1):

RRCP process version (6070025)

Ip Addr = 10.30.57.136 (169752968) network = 10.30.57.136 recvAddr = 0.0.0.0 (INADDR_ANY) MCRecvPort = 37002 (0x908a) MCSendToPort = 37001 (0x9089) PPPort = 37000 (0x9088) MCSendFromPort = (0) not configured devicePath = /tmp/.rrcp/source.0.rrmp packetSize = 3100 maxPktPoolSize = 200000 pktPoolLimitHigh = 190000 pktPoolLimitLow = 180000 shuffleTolerance = 1024 userQLimit = 65535 tdata = 4 ndata = 7 nrreq = 3 trreq = 4 twait = 2 nmissing = 128 tbchold = 30 tpphold = 29 nackDelayTime = 20 bitmapFilter = 0 logger.level = 3 logger.file = /rmds/log/rrcpd/adh_0_source_rrcpdl.log logger.maxSize = 52428800 logger.maxSwapfiles = 5

useIpMulticast = True ipMultTTL = 16 network = net-feed interface = 10.30.57.136 sendMultAddress = 239.254.0.168 recvMultAddress = 239.254.0.169 hsmInterface = 10.30.57.136 hsmMultAddress = 239.254.0.117 hsmPort = 30101 hsmInterval = 1 overflowMsgDump-oldest = 1 maxUsers = 10 recvPortLow = 0 recvPortHigh = 0 udpSendBufSize = 524288 udpRecvBufSize = 524288 nackDelayTime = 20 ackPackingRatio = 10 weightPPRetransSent = 1 weightPPRetransRcvd = 1 weightBCRRequestSent = 1 weightBCRRequestRcvd = 1 congestionHiWaterMark = 50 congestionLowWaterMark = 15 congestionEvaluationInterval = 5 sessionProps = (0x00000001) congestionControl = enabled switchReorderFix = disabled multiThreadEngine = disabled clock tick = 100 --------

And some statistics :

------- RRCP Statistics (IpAddr: 10.30.57.136 (169752968) source mode, device /tmp/.rrcp/source.1):

--- Fri Mar 8 11:12:26 2019

Total pkts sent: 14937491 Rxmt'd PP pkts sent: 2191 BC pkts sent: 14931213 Unack'd PP pkts: 283 PP pkts sent: 6278 RXMTREQPP pkts rcvd: 185 Total pkts rcvd: 920570 RXMTREQPP pkts sent: 25 BC pkts rcvd: 912267 Msgs from users: 14496073 PP pkts rcvd: 8303 BC msgs from users: 14493088 BC DATA pkts sent: 14493088 PP msgs from users: 2985 PP DATA pkts sent: 2985 Msgs to users: 433024 BC DATA pkts rcvd: 430193 DATA msgs to users: 432534 PP DATA pkts rcvd: 2341 BC DATA msgs to users: 430193 ACK pkts sent: 1077 PP DATA msgs to users: 2341 ACKs sent: 2341 STATUS msgs to users: 490 ACK pkts rcvd: 843 Bad pkts/from user: 0 ACKs rcvd: 2672 Bad pkts/from net: 0 PP DATA pkts ackd by wndw: 30 Discards/bad opcode: 0 RXMTREQs staged: 0 Discards/old BC: 0 RXMTREQs canceled: 0 Discards/old PP: 0 RXMTREQ pkts sent: 0 Discards/rxmt'd PP: 0 RXMTREQs sent: 0 Msgs filtered out: 0 Rxmt'd RXMTREQ pkts sent: 0 Loop Msg filtered out: 0 Rxmt'd RXMTREQs sent: 0 BC msgs misordered: 0 Rxmt'd BC pkts rcvd: 0 PP msgs misordered: 0 DISCARD pkts rcvd: 0 Lost data/BC gaps msgs: 0 RXMTREQ pkts rcvd: 1279 Lost data/BC packets: 0 RXMTREQs rcvd: 1819 Lost data/PP gaps msgs: 25 Rxmt'd RXMTREQ pkts rcvd: 3655 Lost data/PP packets: 26 Rxmt'd RXMTREQs rcvd: 5457 Lost data/node resync: 0 Rxmt'd BC pkts sent: 2137 Lost data/msg dscrd'd: 0 DISCARD pkts sent: 0 Lost data/incmplt msg: 0 NULL pkts sent: 435988 Lost data/user Q overflow: 0 NULL pkts rcvd: 482074 Pkt buffers in use: 6 Heartbeats sent: 0 Msg buffers in use: 5 Heartbeats rcvd: 0 Total bytes sent: 2156837849 Rxmt'd PP pkts rcvd: 0 Total bytes recv: 45516739

-----

We checked the server (cables, CRC errors, soft errors), and cannot see any, can you please help to understand why this is failing ?

Even if there is a very small trafic onto this server pair, it looks like RRCPd lost a lot of packets and this, regularly. I am not sure if there is a coincidence with the fact that the ATS "listens" to the ADH hotstandby state and maybe it creates trafic, I really don't know.

Any help will be appreciated ! Thanks, Julien

(ps : I am afraid that the formatting of the statistics will not be helpful here , but it displays well in the support ticket).

elektronrefinitiv-realtimetrepinfrastructurepacket-loss
icon clock
10 |1500

Up to 2 attachments (including images) can be used with a maximum of 5.0 MiB each and 10.0 MiB total.

1 Answer

Upvotes
Accepted
zoya faberov avatar image
23k ●22 ●9 ●14

Hello @julien.dominici,

Having read and re-read your question, it appears that it is not related to any of Refinitiv APIs.

My understanding is, that, having just updated your deployed TREP infrastructure with the latest components, you are concerned with the packet loss, that you are observing.

This forum is dedicated to questions, and general discussion, of Refinitiv APIs.

The moderators of this forum are API experts, and are not equipped to suggest, guide, or contribute meaningfully toward the resolution of the issue that you are facing.

Instead, we suggest that you contact your Refinitiv account team, to discuss the available options to proceed, enabling you to obtain the help that you need, the best path toward the resolution.

To give them a heads-up,I will also copy them on this question,

-AHS

icon clock
10 |1500

Up to 2 attachments (including images) can be used with a maximum of 5.0 MiB each and 10.0 MiB total.

Click below to post an Idea Post Idea