API: The server_id does not change even when restarted. 服务重启 server ID 标识没有改变

[xiangzi1]

Note: Please read FAQ before file an issue, see #2716

Description

Please description your issue here

1. SRS Version: srs5.0.100

2. SRS Log:

Start to run service
[2023-01-11 15:51:51.448][INFO][21][1i3bs59v] XCORE-SRS/5.0.100(Bee)
[2023-01-11 15:51:51.449][INFO][21][1i3bs59v] config parse complete
[2023-01-11 15:51:51.449][INFO][21][1i3bs59v] write log to console
[2023-01-11 15:51:51.449][INFO][21][1i3bs59v] SRS/5.0.100(Bee), MIT
[2023-01-11 15:51:51.449][INFO][21][1i3bs59v] authors: Winlin<winlin@vip.126.com> ZhaoWenjie<zhaowenjie@tal.com> ShiWei<shiwei05@kuaishou.com> XiaoZhihong<hondaxiao@tencent.com> WuPengqiang<pengqiang.wpq@alibaba-inc.com> XiaLixin<xialixin@kanzhun.com> LiPeng<mozhan.lp@alibaba-inc.com> ChenGuanghua<jinxue.cgh@alibaba-inc.com> ChenHaibo<nmgchenhaibo@foxmail.com> Johzzy<hellojinqiang@gmail.com> Zhouxiaojun2008<279686030@qq.com> Pengfei.ma<pengfei.ma.chn@outlook.com> ZhangJunqin<zhangjunqin@jd.com> and https://github.com/ossrs/srs/blob/develop/trunk/AUTHORS.md#contributors
[2023-01-11 15:51:51.449][INFO][21][1i3bs59v] cwd=/usr/local/srs, work_dir=./, build: 2022-12-01 15:19:56, configure: --gb28181=on --sanitizer-static=on, uname: Linux buildkitsandbox 5.15.0-1023-azure #29~20.04.1-Ubuntu SMP Wed Oct 26 19:18:25 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux, osx: 0, env: 0, pkg: dcis
[2023-01-11 15:51:51.449][INFO][21][1i3bs59v] configure detail: --prefix=/usr/local/srs --config=conf/srs.conf --hls=on --hds=off --dvr=on --ssl=on --https=on --ssl-1-0=off --ssl-local=off --sys-ssl=off --transcode=on --ingest=on --stat=on --http-callback=on --http-server=on --stream-converter=on --http-api=on --utest=off --cherrypy=off --srt=on --rtc=on --gb28181=on --simulator=off --cxx11=on --cxx14=off --backtrace=on --ffmpeg-fit=on --nasm=on --srtp-nasm=on --clean=on --gperf=off --gmc=off --gmd=off --gmp=off --gcp=off --gprof=off --static=off --shared-st=off --shared-srt=off --shared-ffmpeg=off --log-verbose=off --log-info=off --log-trace=on --log-level_v2=on --gcov=off --debug=off --debug-stats=off --cross-build=off --sanitizer=on --cygwin64=off --single-thread=off --cc=gcc --cxx=g++ --ar=ar --ld=ld --randlib=randlib
[2023-01-11 15:51:51.449][INFO][21][1i3bs59v] srs checking config...
[2023-01-11 15:51:51.450][INFO][21][1i3bs59v] ips, iface[0] eth0 ipv4 0x11043 172.17.0.2
[2023-01-11 15:51:51.450][INFO][21][1i3bs59v] devices, intranet eth0 172.17.0.2
[2023-01-11 15:51:51.450][WARN][21][1i3bs59v][22] stats network use index=0, ip=172.17.0.2, ifname=eth0
[2023-01-11 15:51:51.450][WARN][21][1i3bs59v][22] stats disk not configed, disk iops disabled.
[2023-01-11 15:51:51.450][INFO][21][1i3bs59v] write log to console
[2023-01-11 15:51:51.451][INFO][21][1i3bs59v] features, rch:on, dash:on, hls:on, hds:off, srt:on, hc:on, ha:on, hs:on, hp:on, dvr:on, trans:on, inge:on, stat:on, sc:on
[2023-01-11 15:51:51.451][INFO][21][1i3bs59v] SRS on amd64 x86_64, conf:conf/docker.conf, limit:1000, writev:1024, encoding:little-endian, HZ:100
[2023-01-11 15:51:51.451][INFO][21][1i3bs59v] mw sleep:350ms. mr enabled:on, default:0, sleep:350ms
[2023-01-11 15:51:51.451][INFO][21][1i3bs59v] gc:on, pq:30000ms, cscc:[0,16), csa:on, tn:on(may hurts performance), ss:auto(guess by merged write)
[2023-01-11 15:51:51.451][INFO][21][1i3bs59v] system default latency(ms): mw(0-350) + mr(0-350) + play-queue(0-30000)
[2023-01-11 15:51:51.451][WARN][21][1i3bs59v][22] SRS/5.0.100 is not stable
[2023-01-11 15:51:51.451][INFO][21][1i3bs59v] write pid=21 to ./objs/srs.pid success!
[2023-01-11 15:51:51.451][INFO][21][1i3bs59v] Thread #1(primordial): init name=srs-master-1, interval=5000ms
[2023-01-11 15:51:51.451][INFO][21][1i3bs59v] Pool: Start threads primordial=1, hybrids=1 ok
[2023-01-11 15:51:51.452][INFO][21][g862ifw7] Thread #2: run with tid=22, entry=0x60c000000400, label=hybrid, name=srs-hybrid-2
[2023-01-11 15:51:51.467][INFO][21][g862ifw7] TencentCloud CLS is disabled
[2023-01-11 15:51:51.467][INFO][21][g862ifw7] TencentCloud APM is disabled
[2023-01-11 15:51:51.473][INFO][21][g862ifw7] fingerprint=D0:63:46:D7:58:01:CE:ED:7D:04:7F:F3:BF:68:EA:2B:68:39:29:3A:7F:64:18:BC:19:29:C8:CF:92:13:06:62
[2023-01-11 15:51:51.473][INFO][21][g862ifw7] CircuitBreaker: enabled=1, high=2x90, critical=1x95, dying=5x99
[2023-01-11 15:51:51.474][INFO][21][g862ifw7] http flv live stream, vhost=__defaultVhost__, mount=[vhost]/[app]/[stream].flv
[2023-01-11 15:51:51.474][INFO][21][g862ifw7] http: root mount to ./objs/nginx/html
[2023-01-11 15:51:51.474][INFO][21][g862ifw7] server main cid=g862ifw7, pid=21, ppid=1, asprocess=0
[2023-01-11 15:51:51.475][INFO][21][g862ifw7] RTMP listen at tcp://0.0.0.0:1935, fd=9
[2023-01-11 15:51:51.475][INFO][21][g862ifw7] HTTP-API listen at tcp://0.0.0.0:1985, fd=10
[2023-01-11 15:51:51.475][INFO][21][g862ifw7] HTTP-Server listen at tcp://0.0.0.0:8080, fd=11
[2023-01-11 15:51:51.475][INFO][21][g862ifw7] signal installed, reload=1, reopen=10, fast_quit=15, grace_quit=3
[2023-01-11 15:51:51.476][INFO][21][g862ifw7] http: api mount /console to ./objs/nginx/html/console
[2023-01-11 15:51:51.477][INFO][21][g862ifw7] rtc listen at udp://0.0.0.0:8000, fd=12
[2023-01-11 15:51:51.477][INFO][21][599u6508] Hybrid cpu=0.00%,38MB
[2023-01-11 15:51:51.477][WARN][21][29404uj8][22] use private address as ip: 172.17.0.2, ifname=eth0
[2023-01-11 15:51:51.477][INFO][21][29404uj8] Startup query id=vid-gv865k1, session=vid-200j0qy, eip=172.17.0.2, wait=300s
[2023-01-11 15:51:51.477][INFO][21][o9ox3e2w] TCP: connection manager run, conns=0
[2023-01-11 15:51:51.478][WARN][21][g862ifw7][11] enable auto reload for docker
[2023-01-11 15:51:51.478][INFO][21][g862ifw7] auto reload watching fd=13, watch=1, file=conf
[2023-01-11 15:51:51.478][INFO][21][6u0408e3] GB: connection manager run, conns=0
[2023-01-11 15:51:51.478][INFO][21][o45my81t] SRT: connection manager run, conns=0
[2023-01-11 15:51:51.478][INFO][21][kc4j60h1] UDP #12 LISTEN at 0.0.0.0:8000, SO_SNDBUF(default=212992, expect=10485760, actual=425984, r0=0), SO_RCVBUF(default=212992, expect=10485760, actual=425984, r0=0)
[2023-01-11 15:51:51.478][INFO][21][gl439156] RTC: connection manager run, conns=0
[2023-01-11 15:51:51.488][INFO][21][hx039xv0] HTTP #0 10.67.4.32:59777 GET http://10.67.65.17:1985/api/v1/vhosts/?callback=angular.callbacks._2c, content-length=-1
[2023-01-11 15:51:51.545][INFO][21][hx039xv0] TCP: before dispose resource(HttpConn)(0x607000004f50), conns=1, zombies=0, ign=0, inz=0, ind=0
[2023-01-11 15:51:51.545][WARN][21][hx039xv0][104] client disconnect peer. ret=1007
[2023-01-11 15:51:51.545][INFO][21][o9ox3e2w] TCP: clear zombies=1 resources, conns=1, removing=0, unsubs=0
[2023-01-11 15:51:51.545][INFO][21][hx039xv0] TCP: disposing #0 resource(HttpConn)(0x607000004f50), conns=1, disposing=1, zombies=0
[2023-01-11 15:51:51.690][INFO][21][49w770rp] HTTP #0 10.67.4.32:59778 GET http://10.67.65.17:1985/api/v1/summaries, content-length=-1
[2023-01-11 15:51:51.734][INFO][21][49w770rp] TCP: before dispose resource(HttpConn)(0x6070000067d0), conns=1, zombies=0, ign=0, inz=0, ind=0
[2023-01-11 15:51:51.734][WARN][21][49w770rp][104] client disconnect peer. ret=1007
[2023-01-11 15:51:51.734][INFO][21][o9ox3e2w] TCP: clear zombies=1 resources, conns=1, removing=0, unsubs=0
[2023-01-11 15:51:51.734][INFO][21][49w770rp] TCP: disposing #0 resource(HttpConn)(0x6070000067d0), conns=1, disposing=1, zombies=0
[2023-01-11 15:51:52.518][INFO][21][1k7q1en9] HTTP #0 10.67.4.32:59780 GET http://10.67.65.17:1985/api/v1/summaries, content-length=-1
[2023-01-11 15:51:52.562][INFO][21][1k7q1en9] TCP: before dispose resource(HttpConn)(0x607000006d10), conns=2, zombies=0, ign=0, inz=0, ind=0

3. SRS Config:

# docker config for srs.
# @see full.conf for detail config.

listen              1935;
max_connections     1000;
# For docker, please use docker logs to manage the logs of SRS.
# See https://docs.docker.com/config/containers/logging/
srs_log_tank        console;
daemon              off;
http_api {
    enabled         on;
    listen          1985;
    raw_api {
        enabled             on;
        allow_reload        on;
    }
}
http_server {
    enabled         on;
    listen          8080;
    dir             ./objs/nginx/html;
}
rtc_server {
    enabled on;
    listen 8000;
    # @see https://github.com/ossrs/srs/wiki/v4_CN_WebRTC#config-candidate
    candidate 10.67.65.17;
}
vhost __defaultVhost__ {
    http_hooks {
        enabled         on;
        on_close        http://xxxxx:3001/api/v1/clients;
        on_play         http://xxxxxx:3001/api/v1/sessions;
        on_stop         http://xxxxxx:3001/api/v1/sessions;
    }
    http_remux {
        enabled     on;
        mount       [vhost]/[app]/[stream].flv;
    }
    rtc {
        enabled     on;
        # @see https://github.com/ossrs/srs/wiki/v4_CN_WebRTC#rtmp-to-rtc
        rtmp_to_rtc on;
        # @see https://github.com/ossrs/srs/wiki/v4_CN_WebRTC#rtc-to-rtmp
        rtc_to_rtmp off;
        stun_timeout 5;
        stun_strict_check on;
    }
}

Replay

Please describe how to replay the bug?

Step 1: Start the SRS service.

API call: http://10.67.65.17:1985/api/v1/server
Response: {

    "code": 0,
    "server": "vid-gv865k1",
    "urls": {
        "versions": "the version of SRS",
        "summaries": "the summary(pid, argv, pwd, cpu, mem) of SRS",
        "rusages": "the rusage of SRS",
        "self_proc_stats": "the self process stats",
        "system_proc_stats": "the system process stats",
        "meminfos": "the meminfo of system",
        "authors": "the license, copyright, authors and contributors",
        "features": "the supported features of SRS",
        "requests": "the request itself, for http debug",
        "vhosts": "manage all vhosts or specified vhost",
        "streams": "manage all streams or specified stream",
        "clients": "manage all clients or specified client, default query top 10 clients",
        "raw": "raw api for srs, support CUID srs for instance the config",
        "clusters": "origin cluster server API",
        "perf": "System performance stat",
        "tcmalloc": "tcmalloc api with params ?page=summary|api"
    },
    "tests": {
        "requests": "show the request info",
        "errors": "always return an error 100",
        "redirects": "always redirect to /api/v1/test/errors",
        "[vhost]": "http vhost for http://error.srs.com:1985/api/v1/tests/errors"
    }
}

Step 2: Restart the SRS service.

SRS container restart.

Step 3: Call the server query interface again.

Call the interface: http://10.67.65.17:1985/api/v1/server
Response: {
    "code": 0,
    "server": "vid-gv865k1",
    "urls": {
        "versions": "the version of SRS",
        "summaries": "the summary(pid, argv, pwd, cpu, mem) of SRS",
        "rusages": "the rusage of SRS",
        "self_proc_stats": "the self process stats",
        "system_proc_stats": "the system process stats",
        "meminfos": "the meminfo of system",
        "authors": "the license, copyright, authors and contributors",
        "features": "the supported features of SRS",
        "requests": "the request itself, for http debug",
        "vhosts": "manage all vhosts or specified vhost",
        "streams": "manage all streams or specified stream",
        "clients": "manage all clients or specified client, default query top 10 clients",
        "raw": "raw api for srs, support CUID srs for instance the config",
        "clusters": "origin cluster server API",
        "perf": "System performance stat",
        "tcmalloc": "tcmalloc api with params ?page=summary|api"
    },
    "tests": {
        "requests": "show the request info",
        "errors": "always return an error 100",
        "redirects": "always redirect to /api/v1/test/errors",
        "[vhost]": "http vhost for http://error.srs.com:1985/api/v1/tests/errors"
    }
}

Expect

After the service restarts, the serverID changes.
Please describe your expectation.

TRANS_BY_GPT3

[xiangzi1]

I tried the latest SRS container service, but the SRS serverID remained unchanged after restarting. The command to start the Docker container is: docker run --restart=always -itd -p 1935:1935 -p 1985:1985 -p 8081:8080 --name srs5.128 srs:v5.0.128.

TRANS_BY_GPT3

[winlinvip]

The server_id is used to identify the machine, so it should not change when restart.

To identify the restart of service, we need another one named service_id:

1. Use Prometheus Exporter, a server, pid and service label will be added to srs_build_info, for example:

# HELP srs_build_info A metric with a constant '1' value labeled by build_date, version from which SRS was built.
# TYPE srs_build_info gauge
srs_build_info{"server":"vid-2566t9t",service="ud72id23d",pid="2846"} 1

3. Each API response, should return pid and service, for example, http://localhost:1985/api/v1/versions:

{
  "code": 0,
  "server": "vid-2566t9t",
  "service": "ud72id23d",
  "pid": 2846,
  "data": {
    "major": 6,
    "minor": 0,
    "revision": 14,
    "version": "6.0.14"
  }
}