Skip to content

Indy Node crashes and closes its listener on heavy load or load for longer period #1868

Open
@SuryaSHalwasia

Description

@SuryaSHalwasia

I am using Indy Node in Ubuntu 20 outside docker. In docker when I run the load test it doesn't crash as it does in Ubuntu 20.
This is the error it gives me

2024-07-05 17:19:08,181|WARNING|stacks.py|Node4C: connections number fell below the limit! Actual: 347, limit: 400.
2024-07-05 17:19:08,181|WARNING|stacks.py|Node4C: client stack restart is not needed anymore.
2024-07-05 17:19:08,878|WARNING|stacks.py|Node4C: connections limit reached! Actual: 410, limit: 400.
2024-07-05 17:19:08,878|WARNING|stacks.py|Node4C: connections limit reached but too few time spent since client stack start, restart it later.
2024-07-05 17:19:15,063|INFO|ordering_service.py|Node4:1 set last ordered as (0, 21)
2024-07-05 17:19:15,071|INFO|ordering_service.py|Node4:1 ordered batch request, view no 0, ppSeqNo 21, ledger 1, state root None, txn root None, audit root None, requests ordered 532, discarded 0
2024-07-05 17:19:15,075|INFO|ordering_service.py|Node4:1 set last ordered as (0, 22)
2024-07-05 17:19:15,081|INFO|ordering_service.py|Node4:1 ordered batch request, view no 0, ppSeqNo 22, ledger 1, state root None, txn root None, audit root None, requests ordered 182, discarded 0
2024-07-05 17:19:15,081|INFO|ordering_service.py|Node4:1 set last ordered as (0, 23)
2024-07-05 17:19:15,086|INFO|ordering_service.py|Node4:1 ordered batch request, view no 0, ppSeqNo 23, ledger 1, state root None, txn root None, audit root None, requests ordered 473, discarded 0
2024-07-05 17:19:15,768|INFO|ordering_service.py|Node4:1 set last ordered as (0, 24)
2024-07-05 17:19:15,769|INFO|ordering_service.py|Node4:1 ordered batch request, view no 0, ppSeqNo 24, ledger 1, state root None, txn root None, audit root None, requests ordered 37, discarded 0
2024-07-05 17:19:24,202|INFO|ordering_service.py|Node4:0 set last ordered as (0, 16)
2024-07-05 17:19:25,005|INFO|ordering_service.py|Node4:0 ordered batch request, view no 0, ppSeqNo 16, ledger 1, state root J3kBWhVEkrhbtwhw7tuB1Y78o49JMSfoUebFvX96xAcc, txn root B9TH7NwF4M6fw6Wk6qYJRBX1VTJjnvKJZPfTSCFGiGPU, audit root 56dnDyFMkxPDTr5xNpTvzfbYFPhhdkDePSGjz5TBuxer, requests ordered 1000, discarded 0
2024-07-05 17:19:33,007|NOTIFICATION|looper.py|Looper shutting down now...
2024-07-05 17:19:33,007|INFO|motor.py|Node4 changing status from started to stopping
2024-07-05 17:19:33,007|INFO|node.py|node Node4 current stats

node inbox size : 0
client inbox size : 0
age (seconds) : 155.4772744178772
next check for reconnect: 6.888445964999988
node connections : {'Node3', 'Node2', 'Node1'}
f : 1
master instance : 0
replicas : 2
view no : 0
rank : 3
msgs to replicas : 7
action queue : 0 140633492501984
action queue stash : 4 140633492502080
2024-07-05 17:19:33,008|INFO|node.py|Node4 reseting...
2024-07-05 17:19:33,011|NOTIFICATION|zstack.py|stack Node4 closing its listener
2024-07-05 17:19:33,012|NOTIFICATION|zstack.py|stack Node4 stopped
2024-07-05 17:19:33,012|NOTIFICATION|zstack.py|stack Node4C closing its listener

And all the nodes crash. My specs are quite good in each of the ubuntu instance. 16GB RAM and 8 CPU.
The dockerized version of indy node is half of it and doesn't crash like this.
What should I do? Thanks

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions