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 <[email protected]>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 <[email protected]> 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 > > [email protected] > > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > > >
_______________________________________________ ceph-users mailing list [email protected] http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
