Hi, John,
Do you mean this error
text format parsing error in line 8: invalid metric name in comment
I install ceph from rpm, but I will try to patch it.
many thanks
when using prometheus 1.7.1 instead of 2.0.0-beta, the "no token found error" disappeared.
Maybe prometheus 2.0.0-beta still has bugs about it.
John Spray <jspray@xxxxxxxxxx>于2017年8月30日周三 下午7:34写道：
On Wed, Aug 30, 2017 at 3:47 AM, shawn tim <tontinme@xxxxxxxxx> wrote:
>
>
> Hello,
> I just want to try prometheus plugin of ceph-mgr.
> Following this doc(http://docs.ceph.com/docs/master/mgr/prometheus/). I get
> output like
>
> [root@ceph01 ~]# curl localhost:9283/metrics/ | head
> % Total % Received % Xferd Average Speed Time Time Time
> Current
> Dload Upload Total Spent Left
> Speed
> 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:--
> 0
> # HELP mds_mem_dir_ Directories closed
> # TYPE mds_mem_dir_ counter
> mds_mem_dir_{daemon="mds.ceph01"} 0.0
> # HELP throttle_msgr_dispatch_throttler_radosclient_max Max value for
> throttle
> # TYPE throttle_msgr_dispatch_throttler_radosclient_max gauge
> throttle_msgr_dispatch_throttler_radosclient_max{daemon="rgw.ceph01"}
> 104857600.0
> # HELP mds_mem_dir+ Directories opened
> # TYPE mds_mem_dir+ counter
> mds_mem_dir+{daemon="mds.ceph01"} 12.0
> .
I notice you have some MDS metrics in there, which cause some invalid
output -- can you try this fix?
https://github.com/ceph/ceph/pull/17318
The "no token found stuff" is not familiar to me: when testing the
prometheus plugin I've always just been using it without any
authentication.
John
> It seems everythins was OK. But when prometheus connect it, I got errors:
>
> WARN[0002] append failed err="no token
> found" source="scrape.go:648" target="{__address__="10.10.0.125:9283",
> __metrics_path__="/metric
> s", __scheme__="http", group="production", instance="10.10.0.125:9283",
> job="ceph"}"
> WARN[0007] append failed err="no token
> found" source="scrape.go:648" target="{__address__="10.10.0.125:9283",
> __metrics_path__="/metric
> s", __scheme__="http", group="production", instance="10.10.0.125:9283",
> job="ceph"}"
> WARN[0012] append failed err="no token
> found" source="scrape.go:648" target="{__address__="10.10.0.125:9283",
> __metrics_path__="/metric
> s", __scheme__="http", group="production", instance="10.10.0.125:9283",
> job="ceph"}"
> ^CWARN[0014] Received SIGTERM, exiting gracefully...
> source="main.go:340"
>
> My prometheus config yml is
>
> - job_name: 'ceph mgr'
> scrape_interval: 5s
> # metrics_path: /metrics/
> # uncomment this still get error (Get http://10.10.0.125:9283/metrics:
> net/http: invalid header field value "Bearer ..." for key Authorization )
> # bearer_token_file: /etc/ceph/ceph.mgr.ceph01.keyring
> # bearer_token: AQD6359ZCMK3DxAAyJYErZayMX/CRZyiyk/UGg==
>
> static_configs:
> - targets: ['10.10.0.125:9283']
> labels:
> group: 'production'
>
>
> My ceph version is
>
> [root@ceph01 ~]# ceph --version
> ceph version 12.1.4 (a5f84b37668fc8e03165aaf5cbb380c78e4deba4) luminous (rc)
>
> My prometheus version is
>
> [root@prom01 prometheus-2.0.0-beta.2.linux-amd64]# ./prometheus --version
> prometheus, version 2.0.0-beta.2 (branch: HEAD, revision:
> a52f082939a566d5269671e98be06fc6bdf61d09)
> build user: root@41a0740ea598
> build date: 20170818-08:16:50
> go version: go1.8.3
>
> Is there any steps I missed
>
> Thanks for help !
>
>
>
> _______________________________________________
> ceph-users mailing list
> ceph-users@xxxxxxxxxxxxxx
> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
>
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
