'Reporting' gui page reporting jails that have been deleted

[tophee]

Hi Guys,

I upgraded to 9.1.0-Release the other day and got into a pickle with my jails/plugins.

I was running 8.3.1 since shortly after it was released and I had minidlna and firefly running perfectly.
I upgraded and messed up my jail/plugin upgrade and realised that I should have run the upgrade script from shell. Once I did that I was able to delete the old jails and plugins (I wanted a fresh start on them).

My 8.3.1 system had a jail and software volume setup for them. However since deleting these volumes, they still appear in the 'Reporting' page of FreeNAS. As I had made several attempts to re-setup my plugins, I have ended up with several 'volumes' being reported in gui that don't exist. Each entry reports the space as "nan B min". As shown below:

How does one go about removing these entries?

Thanks.

 

[dlavigne]

Not sure how to remove them, but it looks like a bug that should be reported at support.freenas.org. Do a search first to make sure that this has not already been reported.

 

[emk2203]

I am also affected by this bug - it seems to be more than just a single occurrence.

In the closed ticket #279 (no gui way to delete and reset reporting data) is a recommendation by gcooper to use

Code:

/usr/local/etc/rc.d/collectd stop && find /var/db/collectd/rrd/localhost/* -type d | xargs rm -Rf && rm /data/rrd_dir.tar.bz2 && /usr/local/etc/rc.d/collectd start && python /root/graph.py

to reset, but this was 2 years ago, and the last part is out-of-date and gives

Code:

python: can't open file '/root/graph.py': [Errno 2] No such file or directory

.

A reboot fixes this, and everything is back to normal then (all graphs there, no stale graphs, everything working).

Would be nice to start graph.py from the CLI, though.

 

[Tom7320]

Sorry but I still don't understand. I have the same problem. How do I reset the Reportings??? What do I have to do?

THX a lot!!

 

[emk2203]

Open the web interface of your FreeNAS box. Second-to-last in the menu on the left is the shell command. Click it and open the shell on the box.

Enter

Code:

/usr/local/etc/rc.d/collectd stop && find /var/db/collectd/rrd/localhost/* -type d | xargs rm -Rf && rm /data/rrd_dir.tar.bz2 && /usr/local/etc/rc.d/collectd start

which clears your reporting settings. Close the shell (type 'exit'), make a reboot from the web page. After the reboot, the reporting should be back to normal.

 

[Tom7320]

THX a lot! It worked for me!

 

[underpickled]

Open the web interface of your FreeNAS box. Second-to-last in the menu on the left is the shell command. Click it and open the shell on the box.

Enter

Code:

/usr/local/etc/rc.d/collectd stop && find /var/db/collectd/rrd/localhost/* -type d | xargs rm -Rf && rm /data/rrd_dir.tar.bz2 && /usr/local/etc/rc.d/collectd start

which clears your reporting settings. Close the shell (type 'exit'), make a reboot from the web page. After the reboot, the reporting should be back to normal.

This deleted the reporting graphs of things that no longer exist, but now that I've rebooted all of the graphs are blank...

EDIT: Sorry, my impatience got the best of me... it just had to populate... working fine and thank you for the solution! I was also getting inaccurate/weird reports about capacities and they seem to be fixed now.

 

[Dusan]

This deleted the reporting graphs of things that no longer exist, but now that I've rebooted all of the graphs are blank...

That solution is actually a bit overkill -- it removes all the graphs and everything starts from zero.
You can go to /var/db/collectd/rrd/localhost/ and selectively remove directories for only those graphs you want to get rid of. Also, no need to restart collectd and you also do not need to delete the .bz2 file -- it is created based on the current graphs on shutdown.

 

[emk2203]

That solution is actually a bit overkill -- it removes all the graphs and everything starts from zero.
You can go to /var/db/collectd/rrd/localhost/ and selectively remove directories for only those graphs you want to get rid of. Also, no need to restart collectd and you also do not need to delete the .bz2 file -- it is created based on the current graphs on shutdown.

Yes, I am fully aware of that. However, my google-fu and limited snooping around in the NAS filesystem didn't turn up much besides the nuke-all solution above.

Where do I find more info on the inner workings of FreeNAS?

 

[Lucien Hercaud]

Hello,
Thanks for the information in this thread. Worked OK for me to delete old entries for a network card I replaced.
Now, I try to add some disk graphs:
- added "LoadPlugin disk" to /usr/local/etc/collectd.conf (my plan is to make it permanent at boot if the reporting works)
- restarted collectd
- I see disk stats RRD's in /var/db/collectd/rrd/localhost/disk-ada*/
But there is none (yet) on the "Reporting page" in the Web interface.
Is there anything I'm missing ?
TIA.

Lucien
___________________________
FreeNAS-9.1.1-RELEASE-x64 (a752d35)
PEGATRON CORPORATION Narra 6 / 8 Gb [ 2*DDR3@1066 MHz]
AMD Athlon(tm) II X2 215 Processor
Gigabit RealTek
6* 3T disks (WD30EZRX-00D8PB0) : raidz1 + spare

 

[dlavigne]

Try a search of "rrds" at https://github.com/freenas/freenas/tree/master/nanobsd. It looks like it is this script that calls the graphs: https://github.com/freenas/freenas/...02d66d358b9257d2b4c13f41/gui/reporting/rrd.py.

 

[Dusan]

- I see disk stats RRD's in /var/db/collectd/rrd/localhost/disk-ada*/
But there is none (yet) on the "Reporting page" in the Web interface.
Is there anything I'm missing ?

It's not that easy. The GUI does not know how to graph the data (colors, areas, lines, labels, ...). You need to add the graphing code to: /usr/local/www/freenasUI/reporting/rrd.py

 

[Lucien Hercaud]

Thanks Dusan.
OK I'll try to go for it.

Cheers and Happy Feasting
Lucien

Gee :
UR fast to answer.

 

[Lucien Hercaud]

Not yet there :-(

I copied rrd.py to /usr/local/www/freenasUI/reporting/rrd.py after " mount -uw / "
Is there a place where python logs errors ?

I still miss something ... Do I need to reboot ? To restart the web server ?
I'm not sure I understand how the new Class is attached to the rest of the Web server.

The code is (not much of my own, just adapted one for SysLoad) :

 

[Dusan]

You need to restart django by running "service django restart". Django/python error message will be in /var/log/messages, web server error messages are in /var/log/nginx-error.log.

 

[Lucien Hercaud]

You need to restart django by running "service django restart". Django/python error message will be in /var/log/messages, web server error messages are in /var/log/nginx-error.log.

Thanks a lot. It works great !
I'll have to take a bit care about numbers' length but that's it.

 

[Lucien Hercaud]

In case someone may want to see the code :

 

[William Grzybowski]

Hi,

Consider forking freenas in github (https://github.com/freenas/freenas) and opening a Pull Request.

Thanks!

 

[Lucien Hercaud]

Hello Will,

I'm not sure I understand your post. Do you want me to do anything ?
Cheers.

 

[William Grzybowski]

Yes, please read https://help.github.com/articles/using-pull-requests