as i understand you, the collectd config is regenerated by omv at every boot, so no matter what changes i'll add to collectd conf, they will be rewritten by ovm on next boot.
am i right?
as i understand you, the collectd config is regenerated by omv at every boot, so no matter what changes i'll add to collectd conf, they will be rewritten by ovm on next boot.
am i right?
may be for USB sticks it's not a solution, but it can work fine with regular scsi (sata) disk that returns correct serial - so it can be a solution in situation like mine, when sata disks get difrent names between reboots!
more over as i understand - some special ID (ID_COLLECTD) can be define via udev rule and uses later in collectd config (although i don't understand how )
"Please note that using an attribute that does not differentiate between the whole disk and its particular partitions (like ID_SERIAL) will result in data about the whole disk and each partition being mixed together incorrectly. In this case, you can use ID_COLLECTD attribute that is provided by contrib/99-storage-collectd.rules udev rule file instead."
I have the following picture - /dev/sdb reports 0.00B in use, while about 10TB is used in reality
while detail button information returns me the correct information:
for /dev/sdb
Label: none uuid: a6d020c2-523e-4dc6-8d3f-c11d89d6bbc4
Total devices 1 FS bytes used 9.67TiB
devid 1 size 16.37TiB used 12.05TiB path /dev/sdb
for /dev/sdc
Label: none uuid: e807e7e6-10a5-453c-afe0-13754571a3db
Total devices 1 FS bytes used 10.47TiB
devid 1 size 14.55TiB used 11.17TiB path /dev/sdc
as well as df returns correct info
/dev/sdc 15T 11T 4.1T 72% /srv/dev-disk-by-uuid-e807e7e6-10a5-453c
/dev/sdb 17T 9.7T 6.5T 60% /srv/dev-disk-by-id-ata-ST18000NM009J
is it possible to fix some how?
PS. /dev/sdb is a host managed zoned device and since omv can't format such a device from webui, it was formated manualy and added manualy to fstab and config.xml.
I found that extra parameters can be used for collectd disk pluging - UdevNameAttr "ID_COLLECTD" (the source - the same info is in here)
but if i edit /etc/collectd/collectd.conf.d/disk.conf to add this parameters - nothig happends
as well no new files appear at /var/lib/rrdcached/db/
LoadPlugin disk
<Plugin disk>
UdevNameAttr "ID_COLLECTD"
IgnoreSelected false
</Plugin>
may be OMV uses other files to store collectd settings
where can i edit the config for collectd?
please advise
PS. Actually i still can't find the udev rule that mentioned it collectd man page ( i.e. "...that is provided by contrib/99-storage-collectd.rules...")
yeh, it seems "just i'm lucky guy"
I use HP Microserver Gen8
every time the host rebooted, my system disk and 2 data disks are named randomly
so, the conslusion - Diskstat plugin can't be used, since the data is not persistent and you never know what disk stats it displays now
am i right?
Hm, i found it Debian Wiki https://wiki.debian.org/Persistent_disk_names
QuoteUnder systemd-udev its not possible to rename block devices, so using NAME=sda in udev rules doesn't work. Instead you must create symlinks using SYMLINK+="mydiska" or something. There are also whole trees of persistent name symlinks created by default now under /dev/disk/by-{id,label,partuuid,path,uuid}.
so the disk name CAN'T be renamed. and since diskstat plugin uses /dev/sd[a-z] naming to collect and display it's data, it can't be used atall.
am i right? or it (diskstat) can be someway directed to use persistant symlinks like /dev/dsk1 or whatevere?
Hm,
could you suggest right rule for this purpose
when i use in my /etc/udev/rules.d/61-myrules.rules
SUBSYSTEM=="block", ATTRS{model}=="TS512MJF110 ", SYMLINK+="flashdrv"
it works - i have permanent name, but that's symlink only
when i use
SUBSYSTEM=="block", ATTRS{model}=="TS512MJF110 ", NAME="sdz"
is doesn't work and i can't find /dev/sdz device
so, statistic confuses avery boot
appreciate a lot for the right command (rule)
thanx a lot
will study the manual
i suppose it would be usefull to add some warning to package description about this issue and probably link to a topic at forum.
it will save time and nerve of people in future
updated OMV 7.4.1
I found that omv-diskstat pluging confuse the data between mounted disks
inspite the info on its tabs (/dev/disk/by-id/ata-XXX, /dev/disk/by-id/ata-YYY ) it operates data like /dev/sda. /dev/sdb, etc...
as a result after reboot disks can change their letter and pluging shows wrong data collected for another disk (see screenshot attached)
Any idea why it happends and how to solve?
Do anybody use USB-stick and Flash-plugin with OMV?
I've got the problem again.
Just installed the fresh copy of OMV to the brand new usb-stick on March 1.
Today my Home nas became unavailable again.
less then 2 weeks of work
Will try the new usb-stick... one more time
thanks for your response
That was output from dmesg...
Tell me please the command to generate error or debug output and i'll do that.
how to check that usb-stick is good?
I did change it recently... But that's flash... i can't be sure...
I'm using OMV since 4.x version.
I have a barebone pc that uses as home NAS
USB stick is used as boot media and HDD as a storage
to save Flash at usb-stick FLASH-plugin is installed...
everything works fine... but just about month or bit more...
then i start receiving a segfault messages
Unfortunately i didn't find how to fix this error...
But if i reinstall it from scratch it works fine... but another month or two...
now a have 5.5.14-1 version installed with all latest update
and week ago i started to receive messages like
"
[ +0.000123] awk[406]: segfault at 7f24d17b9260 ip 00007f24cfc99c4e sp 00007fff064dd790 error 4 in ld-2.28.so[7f24cfc8f000+1e000]
[ +0.000033] Code: 1f 44 00 00 4c 8b 73 08 45 89 f4 49 83 fc 25 0f 84 87 07 00 00 48 8b bd 60 ff ff ff 4c 89 f0 49 8b b7 e8 02 00 00 48 c1 e8 20 <0f> b7 14 47 48 8b bd 58 ff ff ff 48 8d 04 40 48 03 3b 48 89 bd 70
"
does anybody know why it happens?
Is there any way to solve this error without reinstallation?
I love omv so much, but it became terrible for me to use it
thank you
You can run a scheduled job every hour (or whatever you want)
folder2ram -syncall
A sync should happen at shutdown. Did you check if the logfiles are updated during shutdown?
Thanks for your response.
"folder2ram -syncall" works fine and all the info syncs.
for sure i can use some cron rules to recall the command periodically... but is seems to me - that's workaround
What is the default syncing mechanism? I suppose by design folder2ram must sync all the data periodically (once per hour). Unfortunately it doesn't do that in my installation.
so in case of emergency power off it lose all the data since it wasn't synced automatically
thanks in advanced
Hi
i have strange behavior of Flash Memory plugin (latest OMV 5)
it seems to me, it doesn't store the data (sync data) upon reboot or shutdown
1.so, it looks working - i can see (df -h)
folder2ram 970M 13M 958M 2% /var/log
folder2ram 970M 0 970M 0% /var/tmp
folder2ram 970M 600K 970M 1% /var/lib/openmediavault/rrd
folder2ram 970M 0 970M 0% /var/spool
folder2ram 970M 7.7M 963M 1% /var/lib/rrdcached
folder2ram 970M 4.0K 970M 1% /var/lib/monit
folder2ram 970M 1.3M 969M 1% /var/cache/samba
2. the graphs are generated at SystemInfo/Performace Statistics
3. but if i reboot the omv, i get the clean pictures (actually the data that was synced manually by "folder2ram -sync" last time) and statistics starts collecting from the beginning so i'm loosing all the data since manual sync.
Is there any ideas how to fix this issue?
and how to adjust the service (folder2ram) to sync data periodically?
Thanks in advance
hm...
the init script (/etc/init.d/rrdchached) is the same like tin system where rrdcached works fine
how can it be?! at one system the same script starts daemon, at another - no !!!!
so, I found that problem is somewhere in init scripts....
I run rrdcached manually with
/usr/bin/rrdcached -B -F -f 3600 -w 900 -b /var/lib/rrdcached/db/ -j /var/lib/rrdcached/journal/ -p /var/run/rrdcached.pid -l unix:/var/run/rrdcached.sock
and not it works...
So, is there any ideas how to debug the init script?
tried to reinstall rrdcached package with apt-get
no result
is there any ideas hot to debug?