ctr grpc too_many_ping
Created by: ccmeteorljh
cloud地址: http://paddlecloud.baidu-int.com:8088/paddle/jobRunInfo?jobId=job-e6c5b90b17248221&flag=jobs&groupName=paddle&groupId=2b984db1-5c29-5951-b4ee-11da61ccb0a5¤tPage=1¤tKey=1 ctr模型30个trainer,pserver下运行出现下面报错
E0906 15:30:22.608559887 39329 chttp2_transport.cc:1098] Received a GOAWAY with error code ENHANCE_YOUR_CALM and debug data equal to "too_many_pings"
Thu Sep 6 15:30:22 2018[1,12]<stdout>:F0906 15:30:22.608754 39329 grpc_client.cc:295] Send name:[emb@GRAD.trainer_12.pserver_11], ep:[10.89.223.28:8002] meets grpc error:Too many pings
Thu Sep 6 15:30:22 2018[1,12]<stdout>:*** Check failure stack trace: ***
Thu Sep 6 15:30:22 2018[1,12]<stdout>: @ 0x7f570be4201d google::LogMessage::Fail()
Thu Sep 6 15:30:22 2018[1,12]<stdout>: @ 0x7f570be45acc google::LogMessage::SendToLog()
Thu Sep 6 15:30:22 2018[1,12]<stdout>: @ 0x7f570be41b43 google::LogMessage::Flush()
Thu Sep 6 15:30:22 2018[1,12]<stdout>: @ 0x7f570be46fde google::LogMessageFatal::~LogMessageFatal()
Thu Sep 6 15:30:22 2018[1,12]<stdout>: @ 0x7f570c6d3148 paddle::operators::distributed::GRPCClient::Proceed()
Thu Sep 6 15:30:22 2018[1,12]<stdout>: @ 0x7f57198d38a0 execute_native_thread_routine
Thu Sep 6 15:30:22 2018[1,12]<stdout>: @ 0x7f57268631c3 start_thread
Thu Sep 6 15:30:22 2018[1,12]<stdout>: @ 0x7f5725e8b12d __clone
Thu Sep 6 15:30:22 2018[1,12]<stdout>: @ (nil) (unknown)