Open code-brewer opened 3 years ago
Can you also execute server_info
and post the results? Also, what kind of SSD is this? I am assuming EBS, but is it io1, io2, gp2 or gp3?
Thank for your time.
It is io1 type SSD:
There is a partial output of server_state
command in first post, it is exactly the status when gap happen.
Since I have reboot the rippled server, so the server_info
output bellow just reflect the normal status.
sudo docker exec -it 31 rippled server_info
Loading: "/etc/opt/ripple/rippled.cfg"
2021-Aug-26 00:07:56.884747922 UTC HTTPClient:WRN Warning: No peer protocol configured
2021-Aug-26 00:07:56.884804343 UTC HTTPClient:NFO Connecting to 127.0.0.1:5005
{
"result" : {
"info" : {
"build_version" : "1.7.2",
"complete_ledgers" : "65676002-65913098",
"hostid" : "31447bf5c244",
"io_latency_ms" : 1,
"jq_trans_overflow" : "0",
"last_close" : {
"converge_time_s" : 3.001,
"proposers" : 41
},
"load" : {
"job_types" : [
{
"job_type" : "untrustedValidation",
"per_second" : 24
},
{
"avg_time" : 6,
"job_type" : "ledgerRequest",
"peak_time" : 405,
"per_second" : 12
},
{
"job_type" : "untrustedProposal",
"per_second" : 13
},
{
"in_progress" : 2,
"job_type" : "ledgerData",
"peak_time" : 9,
"waiting" : 1
},
{
"in_progress" : 1,
"job_type" : "clientCommand",
"peak_time" : 24,
"per_second" : 34
},
{
"job_type" : "transaction",
"per_second" : 12
},
{
"job_type" : "batch",
"peak_time" : 1,
"per_second" : 11
},
{
"job_type" : "advanceLedger",
"peak_time" : 19,
"per_second" : 4
},
{
"job_type" : "fetchTxnData",
"per_second" : 5
},
{
"job_type" : "trustedValidation",
"peak_time" : 2,
"per_second" : 8
},
{
"in_progress" : 1,
"job_type" : "acceptLedger"
},
{
"job_type" : "trustedProposal",
"per_second" : 18
},
{
"avg_time" : 2,
"job_type" : "heartbeat",
"peak_time" : 5
},
{
"job_type" : "peerCommand",
"peak_time" : 2,
"per_second" : 1323
},
{
"job_type" : "processTransaction",
"per_second" : 12
},
{
"job_type" : "SyncReadNode",
"peak_time" : 39,
"per_second" : 359
},
{
"job_type" : "AsyncReadNode",
"peak_time" : 2,
"per_second" : 1735
},
{
"job_type" : "WriteNode",
"peak_time" : 1,
"per_second" : 435
}
],
"threads" : 6
},
"load_factor" : 1,
"peer_disconnects" : "1",
"peer_disconnects_resources" : "0",
"peers" : 21,
"pubkey_node" : "n9MTg18H5uLJTG2P3P62nDMHkNjDm9BjbGqy12ZiCcP9RpwkTWsd",
"pubkey_validator" : "none",
"server_state" : "full",
"server_state_duration_us" : "135747619612",
"state_accounting" : {
"connected" : {
"duration_us" : "234430924",
"transitions" : 2
},
"disconnected" : {
"duration_us" : "1140696",
"transitions" : 2
},
"full" : {
"duration_us" : "135747619612",
"transitions" : 1
},
"syncing" : {
"duration_us" : "4099507",
"transitions" : 1
},
"tracking" : {
"duration_us" : "40",
"transitions" : 1
}
},
"time" : "2021-Aug-26 00:07:56.886485 UTC",
"uptime" : 135987,
"validated_ledger" : {
"age" : 3,
"base_fee_xrp" : 1e-05,
"hash" : "54890F2B2BADDCEC5B16E3E8C0FEB8B88D4C86478D396CAE7D59AAFECF0C90AA",
"reserve_base_xrp" : 20,
"reserve_inc_xrp" : 5,
"seq" : 65913098
},
"validation_quorum" : 33,
"validator_list" : {
"count" : 1,
"expiration" : "2022-Jan-11 00:00:00.000000000 UTC",
"status" : "active"
}
},
"status" : "success"
}
}
The log during gap happen: rippled_debug.log
Thanks for the report and the additional data @code-brewer. Stand by.
We set up a node use docker image from https://hub.docker.com/r/xrptipbot/rippled/ But it reports ledger gap frequently, maybe two or three days a time, or once per month.
It seem to have relation with IOPS ( This server is aws ec2 with a 350GB SSD of 12500 IOPS ), but I don't sure.
I check with
iostat -x 1
, got:My rippled configure is:
What
server_state
command output is:Any ideas?