Mass collectd, rrdcached Errors in syslog

  • Seeing same errors as in http://phpbb.openmediavault.or…1ea5de8395633b041daeb7a06


    OMV/Debian on /, RAID1 built on btrfs (custom made, not through OMV interface)


    Any solution?


    Thanks in advance.


  • I had the same errors on my system. I have two different Raids (Raid1 for system, Raid5 for data).
    If the devices on the raid 5 (2 LVM Volumes) were unmounted (but formatted) the errors would come every minute, after mounting the drives the errors did go away.


    So if you have unmounted drives or volumes try to mount them.

  • I use btrfs so the filesystem is manually mounted, not via omv. Any possible solution...?


    An ugly hack will do, the logs files ballooned into several GBs taking down the server.


    Thx in advance.

    • Offizieller Beitrag

    Comment out the "LoadPlugin df" section in /etc/collectd/collectd.conf. Then monit restart collectd. This change will be overwritten with certain changes in the web interface or possibly an omv update.

    omv 7.0.4-2 sandworm | 64 bit | 6.5 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.10 | compose 7.1.2 | k8s 7.0-6 | cputemp 7.0 | mergerfs 7.0.3


    omv-extras.org plugins source code and issue tracker - github


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • ok. so not a good solution.


    but here is what I did:

    Code
    monit stop collectd


    what is the downside of this? besides not getting constant information I actually do not need. I mean if I need to monitor I can probably also do it with some other CLI tools.


    Would this be a good work arround until a fix is found or at least until Jessie is out? Or will it break some other thing?


    I tried it only in virtualbox and it seemed to stop the logs and I ahvent' noticed any side effects, but I did this just before going to work so I only had a min or two to do the checks.

  • no it's a df debian bug (explained in another thread).: mass errors in syslog (clean install upgraded to .6)
    solution provided by chrunchbang (http://crunchbang.org/forums/viewtopic.php?id=25901) doesn't solve the collectd issue though it does make the logs look slightly nicer.


    What Aaron suggest works but as he noted

    Zitat

    This change will be overwritten with certain changes in the web interface or possibly an omv update.


    for update it would be an easy fix (e.g. with a script) but changes to web interface....


    So again my question is - is stopping collectd a good workaround? or do I break anything else with that? I use ext4 and as I understand here the collectd is only collecting system data for display.


    another option would be to make that file read only after that change, however again I am not sure what that would mean when

    Zitat

    certain changes in the web interface or possibly an omv update

    appear.

    • Offizieller Beitrag

    You are fine to stop collectd but upgrades or even a reboot (if you just use monit stop) will start it again.

    omv 7.0.4-2 sandworm | 64 bit | 6.5 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.10 | compose 7.1.2 | k8s 7.0-6 | cputemp 7.0 | mergerfs 7.0.3


    omv-extras.org plugins source code and issue tracker - github


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • excellent! I've been checking collectd pages. I am not familiar with code or git... but is this some kind of fix?:
    https://github.com/svn2github/…272c85afc46e6724f5ca4cfd0


    or is this just a proposal for a solution from Openmediavault?


    I will either try to implement your fix in some way (read only config or maybe to get notified if file changes) or maybe just turn it off completely until it gets fixed. The logs stay more or less quiet after this is turned off. I am off to find other tools for monitoring I could use.


    Never expected such issues from Debian stable and it's official packages...

    • Offizieller Beitrag

    That fix is very old (omv 0.6) and was an attempt to fix the problem.


    While it is a bug, it really doesn't hurt much other than logging too much. It is a big deal for a RPi but not really any other system.

    omv 7.0.4-2 sandworm | 64 bit | 6.5 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.10 | compose 7.1.2 | k8s 7.0-6 | cputemp 7.0 | mergerfs 7.0.3


    omv-extras.org plugins source code and issue tracker - github


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • Hi,
    I have a lot of message in log (clean last omv from scratch)

    Code
    Oct 25 22:52:31 omv collectd[2191]: rrdcached plugin: rrdc_update (/var/lib/rrdcached/db/localhost/df-root/df_complex-free.rrd, [1414270351:55327059968.000000], 1) failed with status -1.
    Oct 25 22:52:31 omv collectd[2191]: Filter subsystem: Built-in target `write': Dispatching value to all write plugins failed with status -1.
    Oct 25 22:52:31 omv collectd[2191]: rrdcached plugin: rrdc_update (/var/lib/rrdcached/db/localhost/df-root/df_complex-reserved.rrd, [1414270351:3020947456.000000], 1) failed with status -1.
    Oct 25 22:52:31 omv collectd[2191]: Filter subsystem: Built-in target `write': Dispatching value to all write plugins failed with status -1.
    Oct 25 22:52:31 omv collectd[2191]: rrdcached plugin: rrdc_update (/var/lib/rrdcached/db/localhost/df-root/df_complex-used.rrd, [1414270351:1121165312.000000], 1) failed with status -1.


    Is there any way stop this ?
    Thanks

    • Offizieller Beitrag

    Add a data drive

    omv 7.0.4-2 sandworm | 64 bit | 6.5 proxmox kernel

    plugins :: omvextrasorg 7.0 | kvm 7.0.10 | compose 7.1.2 | k8s 7.0-6 | cputemp 7.0 | mergerfs 7.0.3


    omv-extras.org plugins source code and issue tracker - github


    Please try ctrl-shift-R and read this before posting a question.

    Please put your OMV system details in your signature.
    Please don't PM for support... Too many PMs!

  • Hi!
    I have exactly the same problem:
    Nov 6 18:45:23 NAS collectd[2301]: rrdcached plugin: rrdc_update (/var/lib/rrdcached/db/localhost/df-root/df_complex-free.rrd, [1415295923:6860156928.000000], 1) failed with status -1.
    Nov 6 18:45:23 NAS collectd[2301]: Filter subsystem: Built-in target `write': Dispatching value to all write plugins failed with status -1.
    Nov 6 18:45:23 NAS collectd[2301]: rrdcached plugin: rrdc_update (/var/lib/rrdcached/db/localhost/df-root/df_complex-reserved.rrd, [1415295923:446271488.000000], 1) failed with status -1.
    Nov 6 18:45:23 NAS collectd[2301]: Filter subsystem: Built-in target `write': Dispatching value to all write plugins failed with status -1.


    and yes, I have a data drive...how can this be stopped? Its logging like 40 lines per minute....(and I have an ARM device)


    I also read mass errors in syslog (clean install upgraded to .6) but it seems this issue is still open.


    pls note that I have mounted by EXT4 without ACL, QUOTA (OMV 1.0.29 Partition mount problem) in case its related


    and for the graphs, the only one that doesn't get done is the "disk usage"

  • got it !! :)


    in /etc/collectd/collectd.conf


    the DF section was:


    Code
    LoadPlugin df
    <Plugin df>
    # MountPoint "/"
      IgnoreSelected false
    </Plugin>


    so I uncommented MountPoint and manually pointed to the correct mount in /media...for the data drive that appears in the tab of "system information", "state", "disk usage"

Jetzt mitmachen!

Sie haben noch kein Benutzerkonto auf unserer Seite? Registrieren Sie sich kostenlos und nehmen Sie an unserer Community teil!