Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    pmg
    @pmg7557_twitter

    For node "Philipp3' (end of 1/2, both are ok)

    grep jobScheduler /var/rudder/cfengine-community/outputs/* | grep launched | grep sauve_bd2 | cut -d' ' -f 1
    /var/rudder/cfengine-community/outputs/cf_philippe31601590052_Fri_Oct2_00_07_32_2020_0x7f786a056700:2020-10-01T22:09:21+00:00
    /var/rudder/cfengine-community/outputs/cf_philippe31601762852_Sun_Oct4_00_07_32_2020_0x7f786a056700:2020-10-03T22:08:02+00:00
    /var/rudder/cfengine-community/outputs/cf_philippe31601935626_Tue_Oct6_00_07_06_2020_0x7f786a056700:2020-10-05T22:07:38+00:00
    /var/rudder/cfengine-community/outputs/cf_philippe31602022038_Wed_Oct7_00_07_18_2020_0x7f20cced5700:2020-10-06T22:07:46+00:00

    grep jobScheduler /var/rudder/cfengine-community/outputs/* | grep launched | grep tmp_check | cut -d' ' -f 1
    /var/rudder/cfengine-community/outputs/cf_philippe31601590052_Fri_Oct2_00_07_32_2020_0x7f786a056700:2020-10-01T22:09:21+00:00
    /var/rudder/cfengine-community/outputs/cf_philippe31601762852_Sun_Oct4_00_07_32_2020_0x7f786a056700:2020-10-03T22:08:02+00:00
    /var/rudder/cfengine-community/outputs/cf_philippe31601935626_Tue_Oct6_00_07_06_2020_0x7f786a056700:2020-10-05T22:07:38+00:00
    /var/rudder/cfengine-community/outputs/cf_philippe31602022038_Wed_Oct7_00_07_18_2020_0x7f20cced5700:2020-10-06T22:07:47+00:00

    @ncharles With these logs I can imagine that there are two or three different pbs, persistence of the status info, launch of the jobs and ?. Let me know if you want me to do more tests.
    pmg
    @pmg7557_twitter
    @ncharles I created job2 (tmp_check) for these tests, it does not consume ressources I'll change the opening time from 0-23 to 0-2, it should be easier to analyse the pbs.
    Nicolas Charles
    @ncharles
    Thank you
    vpn1 doesn't answer yet, because last run at 8:55 executed the command, so report will be at next run: 9:55
    Nicolas Charles
    @ncharles
    it should be ok now
    pmg
    @pmg7557_twitter
    @ncharles Yes, ok for job2 as you wrote; but of course still KO for job1 (not ran this night)
    @ncharles For your information when I updated teh bug 1803, I set severity to major as I don't know how to solve it inside rudder, but it's not a big pb as it's easy to scedule the job outside rudder (cron, ...)
    Nicolas Charles
    @ncharles
    perfect
    Alex
    @alex17865_twitter
    Hi guys, I have this problem too: https://issues.rudder.io/issues/18257
    Do you know if a solution exist ?
    Nicolas Charles
    @ncharles
    Hi @alex17865_twitter - we don't have a solution yet, but the workaround is described in the issue - it should help you workaround it
    Alex
    @alex17865_twitter
    I have a better workaround I think. Go in <Rudder_Dir>, and do "git add techniques/* && git commit -a" in ssh.
    But it's a little bit annoying to do this each time ><
    Francois Armand
    @fanf
    @alex17865_twitter what Rudder version? (to update the ticket if still present in 6.1)
    Suvi
    @Suvi8

    Hallo, i have a problem updating the OS with yum update:

    it gives me this error:

    yum update
    Loaded plugins: fastestmirror
    Setting up Update Process
    Determining fastest mirrors
    epel/metalink | 29 kB 00:00

    • base: mirror.init7.net
    • epel: pkg.adfinis-sygroup.ch
    • extras: mirror.init7.net
    • updates: mirror.init7.net
      Rudder_6.0 | 2.9 kB 00:00
      Rudder_6.0/primary_db | 5.7 kB 00:00
      https://download.rudder.io/rpm/6.1/RHEL_6/repodata/repomd.xml: [Errno 14] PYCURL ERROR 22 - "The requested URL returned error: 401 Unauthorized"
      Trying other mirror.
      Error: Cannot retrieve repository metadata (repomd.xml) for repository: Rudder_6.1. Please verify its path and try again
    Its a CentOS 6.10
    Alexis Mousset
    @amousset
    Hi @Suvi8, CentOS 6 requires usingthe private repositories as it is not publicly supported anymore, but only as part of the subscription.
    Alex
    @alex17865_twitter
    @fanf last one on debian 10
    Alexis Mousset
    @amousset
    https://docs.rudder.io/reference/6.1/installation/agent/rhel.html#_installation in the blue info box :
    echo '[Rudder_6.1]
    name=Rudder 6.1
    baseurl=https://LOGIN:PASSWORD@download.rudder.io/rpm/6.1/RHEL_$releasever/
    gpgcheck=1
    gpgkey=https://LOGIN:PASSWORD@download.rudder.io/rpm/rudder_rpm_key.pub' > /etc/yum.repos.d/rudder.repo
    Suvi
    @Suvi8
    @amousset yes, we have a rudder subscription
    Alexis Mousset
    @amousset
    (and replace LOGIN/PASSWORD by your technical account's credentials)
    Suvi
    @Suvi8

    @amousset its already there, when i want to clean the yum cache it shows:

    yum clean all
    Loaded plugins: fastestmirror
    Repository Rudder_6.1 is listed more than once in the configuration

    Alexis Mousset
    @amousset
    Oh indeed I did not read the first message correctly!
    Suvi
    @Suvi8
    yesi can access the repo with username/password in my browser
    Francois Armand
    @fanf
    @alex17865_twitter just to be sure: last 6.1 or 6.0 (sorry to ask, it's to be able to reproduce in your condition, else we won't know for sure if we corrected the pb)
    Alex
    @alex17865_twitter
    @fanf 6.1 :)
    Francois Armand
    @fanf
    @alex17865_twitter thanks, ticket updated accordingly
    _ChezW@m_
    @JrmChP_twitter
    Hey, is there an easy way to change a node uuid post re install, or do i have to delete it from rudder, and accept again ?
    Francois Armand
    @fanf
    @JrmChP_twitter depends of your use case. If you reinstalled rudder agent on the same node, and want a different uuid, you can use rudder agent reinit which will reset everything, create a new uuid, and send an inventory. Still, you will have to delete the old node/uuid and accepte the new one.
    _ChezW@m_
    @JrmChP_twitter
    ok thanks. no way to change the uuid on rudder after an os reinstall for example ?
    Francois Armand
    @fanf
    @JrmChP_twitter I don't know what you want to achieve. If it's an OS reinstall, it is a different node, no? Likely, not the same rules/groupes/etc. What would you like to keep from the other node/uuid ?
    _ChezW@m_
    @JrmChP_twitter
    @fanf juste a reinstall of same os / config on a host that was used for test before production. I was lazy and want to avoid declaration of properties :-)
    Francois Armand
    @fanf
    @JrmChP_twitter perhaps you could have used the old uuid, but there is some subtlilies regarding agent keys restoration, not sure it's documented
    _ChezW@m_
    @JrmChP_twitter
    ok thanks @fanf . I finally juste remove and accept nodes again. Only 3 manual properties to apply. by the way, maybe full copying a property from one node to another may be a cool improvements on future release :-)
    Francois Armand
    @fanf
    @JrmChP_twitter datasources plugin allows to copy properties from an end point
    pmg
    @pmg7557_twitter
    Hi All, I want to retrieve a rudder property for all my nodes. Does somebody know the right ldap request?
    Francois Armand
    @fanf
    @pmg7557_twitter not sure why you want to use LDAP ? That's an internal API, not meant to be public. Don't you want to use API for that ? curl -k -H "X-API-TOKEN: ..." -H "Content-Type: application/json" 'https://..../rudder/api/latest/nodes?include=minimal,properties' | jq '.data.nodes[] (here, you can be much more precise with filter on prop name, etc, see: https://stedolan.github.io/jq/manual/
    for example, to get values for all property names "test" in a json array that can be latter post-processed, you can do that :
    .../rudder/api/latest/nodes?include=minimal,properties' | jq '[.data.nodes[] | {"id":.id, "propTest": .properties | map(select(.name == "test"))}]'
    and the result will looks like:
    image.png
    pmg
    @pmg7557_twitter
    @fanf Thanks François. I thought of ldap because I just wanted to do a one quick one shoot test. I never used the API, I 'll see it later because for the moment curl failed due to lack of certifcat.
    Francois Armand
    @fanf
    @pmg7557_twitter even with -k ?
    @pmg7557_twitter also, you have a local token on server: curl -k -H "X-API-Token: $(cat /var/rudder/run/api-token)" ...
    pmg
    @pmg7557_twitter
    @fanf Ok with -k, very easy to use :-) THANKS. Just to fin a little bug in your filter ( bash: erreur de syntaxe près du symbole inattendu « select » )
    Francois Armand
    @fanf
    @pmg7557_twitter du you have single quotes around the whole expression? '[.data.nodes[] | {"id":.id, "propTest": .properties | map(select(.name == "test"))}]'
    bash should not look into it
    pmg
    @pmg7557_twitter
    @fanf Thanks, it works perfectly :-)
    Francois Armand
    @fanf
    :thumbsup:
    Waccabac
    @Waccabac_twitter
    Hi
    Does Rudder agent get Docker informations ?