[Bug]: [Nightly] Milvus deployment failed because of pods crash

[NicoYuan1986]

Is there an existing issue for this?

• I have searched the existing issues

Environment

- Milvus version: master(0449c74)
- Deployment mode(standalone or cluster):cluster&&standalone
- MQ type(rocksmq, pulsar or kafka):    
- SDK version(e.g. pymilvus v2.0.0rc2):
- OS(Ubuntu or CentOS): 
- CPU/Memory: 
- GPU: 
- Others:

Current Behavior

Milvus deployment failed because of pods crash.
link: https://jenkins.milvus.io:18080/blue/organizations/jenkins/Milvus%20Nightly%20CI(new)/detail/master/167/pipeline/151

[milvus-install : deploy] Failed
task check-install-status has failed: "step-check-status" exited with code 1 (image: "harbor.milvus.io/milvusdb/bitnami/kubectl@sha256:95276c7786df8cc7aef7ed874b56084b1c7047858f82c2db0bca5f5f4eb5d9df"); for logs run: kubectl -n milvus-tekton logs milvus-pytest-6nh45-check-install-status-pod -c step-check-status
[check-install-status : check-status] ----------------Pod Status --------------------------------------------
[check-install-status : check-status] mdk-master-167-py-n-etcd-0                                  1/1     Running            0               13m    10.104.15.158   4am-node20   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-etcd-1                                  1/1     Running            0               13m    10.104.23.153   4am-node27   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-etcd-2                                  1/1     Running            0               13m    10.104.30.24    4am-node38   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-kafka-0                                 2/2     Running            3 (12m ago)     13m    10.104.33.63    4am-node36   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-kafka-1                                 2/2     Running            2 (12m ago)     13m    10.104.15.161   4am-node20   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-kafka-2                                 2/2     Running            2 (12m ago)     13m    10.104.24.253   4am-node29   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-kafka-exporter-556fc7f678-7ztp4         1/1     Running            5 (12m ago)     13m    10.104.6.11     4am-node13   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-milvus-datanode-59cdd5b54b-9ltz4        0/1     CrashLoopBackOff   7 (73s ago)     13m    10.104.5.13     4am-node12   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-milvus-datanode-59cdd5b54b-jrwjd        0/1     CrashLoopBackOff   7 (73s ago)     13m    10.104.6.9      4am-node13   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-milvus-indexnode-86bfbdb68c-77sgj       0/1     CrashLoopBackOff   7 (67s ago)     13m    10.104.6.10     4am-node13   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-milvus-indexnode-86bfbdb68c-7m8gz       0/1     CrashLoopBackOff   7 (70s ago)     13m    10.104.14.184   4am-node18   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-milvus-mixcoord-84b744c9bd-hqljz        0/1     CrashLoopBackOff   7 (109s ago)    13m    10.104.6.12     4am-node13   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-milvus-proxy-7dcdb9c558-4tgmh           0/1     CrashLoopBackOff   7 (86s ago)     13m    10.104.6.8      4am-node13   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-milvus-proxy-7dcdb9c558-ntzvp           0/1     CrashLoopBackOff   7 (70s ago)     13m    10.104.14.185   4am-node18   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-milvus-querynode-5475446485-vs2h9       0/1     CrashLoopBackOff   7 (63s ago)     13m    10.104.14.186   4am-node18   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-milvus-querynode-5475446485-zzxdc       0/1     CrashLoopBackOff   7 (69s ago)     13m    10.104.6.13     4am-node13   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-minio-6f56744d6d-hzrpj                  1/1     Running            0               13m    10.104.30.21    4am-node38   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-zookeeper-0                             1/1     Running            0               13m    10.104.16.155   4am-node21   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-zookeeper-1                             1/1     Running            0               13m    10.104.23.154   4am-node27   <none>           <none>
[check-install-status : check-status] mdk-master-167-py-n-zookeeper-2                             1/1     Running            0               13m    10.104.15.162   4am-node20   <none>           <none>

panic log:

2024-11-03T02:03:28.495306652+08:00 stdout F [2024/11/02 18:03:28.495 +00:00] [ERROR] [components/query_node.go:60] ["QueryNode starts error"] [error="context deadline exceeded"] [stack="github.com/milvus-io/milvus/cmd/components.(*QueryNode).Run\n\t/workspace/source/cmd/components/query_node.go:60\ngithub.com/milvus-io/milvus/cmd/roles.runComponent[...].func1\n\t/workspace/source/cmd/roles/roles.go:129"]
2024-11-03T02:03:28.49747112+08:00 stderr F panic: context deadline exceeded
2024-11-03T02:03:28.497484485+08:00 stderr F
2024-11-03T02:03:28.497489948+08:00 stderr F goroutine 231 gp=0xc001803500 m=18 mp=0xc001800808 [running]:
2024-11-03T02:03:28.497494659+08:00 stderr F panic({0x63eff60?, 0x9efa280?})
2024-11-03T02:03:28.497499328+08:00 stderr F    /go/pkg/mod/golang.org/toolchain@v0.0.1-go1.22.0.linux-amd64/src/runtime/panic.go:779 +0x158 fp=0xc001da3f70 sp=0xc001da3ec0 pc=0x1f82b18
2024-11-03T02:03:28.497502897+08:00 stderr F github.com/milvus-io/milvus/cmd/roles.runComponent[...].func1()
2024-11-03T02:03:28.497507757+08:00 stderr F    /workspace/source/cmd/roles/roles.go:130 +0x128 fp=0xc001da3fe0 sp=0xc001da3f70 pc=0x5cb7ea8
2024-11-03T02:03:28.497512032+08:00 stderr F runtime.goexit({})
2024-11-03T02:03:28.497516923+08:00 stderr F    /go/pkg/mod/golang.org/toolchain@v0.0.1-go1.22.0.linux-amd64/src/runtime/asm_amd64.s:1695 +0x1 fp=0xc001da3fe8 sp=0xc001da3fe0 pc=0x1fc1981
2024-11-03T02:03:28.497520089+08:00 stderr F created by github.com/milvus-io/milvus/cmd/roles.runComponent[...] in goroutine 1
2024-11-03T02:03:28.497524327+08:00 stderr F    /workspace/source/cmd/roles/roles.go:118 +0x129
2024-11-03T02:03:28.497528931+08:00 stderr F
2024-11-03T02:03:28.49753247+08:00 stderr F goroutine 1 gp=0xc0000061c0 m=nil [semacquire]:
2024-11-03T02:03:28.497536373+08:00 stderr F runtime.gopark(0x128f9d24f5fbe04?, 0x120?, 0x0?, 0x0?, 0x120?)
2024-11-03T02:03:28.497550587+08:00 stderr F    /go/pkg/mod/golang.org/toolchain@v0.0.1-go1.22.0.linux-amd64/src/runtime/proc.go:402 +0xce fp=0xc0021c15b8 sp=0xc0021c1598 pc=0x1f8728e
2024-11-03T02:03:28.497561532+08:00 stderr F runtime.goparkunlock(...)

server log: artifacts-milvus-distributed-kafka-mdk-master-167-py-n-167-e2e-logs.tar.gz

Expected Behavior

deploy successfully

Steps To Reproduce

No response

Milvus Log

No response

Anything else?

No response

[NicoYuan1986]

I will try to fix by updating milvus helm chart version to 4.2.18. (It is 4.2.8 now)

[NicoYuan1986]

#37406

[yanliang567]

maybe the same to #37402, the address is null

2024-11-03T02:11:02.374889923+08:00 stdout F [2024/11/02 18:11:02.374 +00:00] [INFO] [rootcoord/root_coord.go:159] ["update rootcoord state"] [state=Abnormal]
2024-11-03T02:11:02.375166299+08:00 stdout F [2024/11/02 18:11:02.375 +00:00] [INFO] [rootcoord/service.go:154] ["RootCoord listen on"] [address="[::]:19530"] [port=19530]
2024-11-03T02:11:02.375205971+08:00 stdout F [2024/11/02 18:11:02.375 +00:00] [INFO] [rootcoord/service.go:196] ["init params done.."]
2024-11-03T02:11:02.375338359+08:00 stdout F [2024/11/02 18:11:02.375 +00:00] [INFO] [etcd/etcd_util.go:52] ["create etcd client"] [useEmbedEtcd=false] [useSSL=false] [endpoints="[localhost:2379]"] [minVersion=1.3]
2024-11-03T02:11:02.375990525+08:00 stdout F [2024/11/02 18:11:02.375 +00:00] [WARN] [datacoord/service.go:95] ["DataCoord fail to create net listener"] [error="listen tcp :19530: bind: address already in use"]

[yanliang567]

@congqixia & @LoveEachDay are working on it.

/assign @congqixia
/unassign

[xiaofan-luan]

why this need a fix?

this is just listen tcp :19530: bind: address already in use

which means the 19530 port has been used by other process

[xiaofan-luan]

don't think there is a issue here

[yanliang567]

The error msg is a bit confusing, but the root cause here is the ip address was parsed to null for an empty config.

[yanliang567]

I can manually deploy milvus now with the pr #37418, but the nightly run still failed in deployment with the same commit. @congqixia do we still need your fix pr above?

[congqixia]

@yanliang567 the root cause shall be the misbehavior of yaml parser. #37418 shall be handling null case for yaml configuration. I shall check why the nightly failed with this patch

[congqixia]

why this need a fix?

this is just listen tcp :19530: bind: address already in use

which means the 19530 port has been used by other process

@xiaofan-luan

if milvus failed to discover a viable ip address and put it in session, other component could never be able to connect to coordinators and vise versa

[congqixia]

The root cause was behavior change of yaml parsing.
Empty yaml file (with # comments, which means length not zero) will report error and make other yaml file fail to work.
#37445 skip EOF error to fix this problem

[congqixia]

It seems that last nightly run started successfully, could you please verify?
/unassign
/assign @NicoYuan1986

[NicoYuan1986]

The issue has been fixed. Thanks for all your help ~ 🌈