On Wed, Aug 30, 2017 at 5:54 PM, shawn tim <tontinme@xxxxxxxxx> wrote: > Hi, John, > Do you mean this error > text format parsing error in line 8: invalid metric name in comment Yes, iirc that's the one you get from the plus/minus signs in MDS metric names. > 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. Huh, weird, I think I was only ever using 2.0. No idea then! John > > 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
 |