Just installed the latest rockstor on my home server using bare metal. The install went smooth, but somehow I added ALL my storage to the rockstor_rockstor pool.
Brief description of the problem
Cannot remove disks from the rockstor_rockstor pool. When I try to run the resize tool to remove a disk, it crashes out with s script error. Is there a way to remove the 2 1tb disks from the pool, or should I use the setup again?
Detailed step by step instructions to reproduce the problem
File “/opt/rockstor/src/rockstor/rest_framework_custom/generic_view.py”, line 41, in _handle_exception
yield
File “/opt/rockstor/src/rockstor/storageadmin/views/pool.py”, line 351, in put
‘as it contains the operating system.’ % pid)
TypeError: %d format: a number is required, not unicode
The error message display code you received when trying to back pedal is actually throwing an error itself (so you only see half of it) but this has been fixed in more recent code than you ran (currently only available in stable channel updates or by manual building):
Which was released in stable channel update version 3.9.2-21 (3rd April). See our GitHub Releases page.
There is probably a ‘clever’ way out but really if you have only just gotten started, ie no data on the system yet, then I’d definitely favour the simple re-install approach. You may even have to wipe the data drives as they will remember being part of a rockstor_rockstor pool. Although you will probably be able to wipe them from within the re-installed system prior to creating a fresh data only pool with them.
Thanks for reporting your finding, I’ve added a link to the issue to point back to your report here so this all helps with highlighting the more pressing needs issues wise.
Hope that helps and let us know how it goes. This thread should be updated when that issue is resolved.
And remember to subscribe to one or other update channels after re-installing as both have many updates available on top of the latest iso.
Thanks for your help, I was able to get my NAS working and serving my household again! Thanks for your help! I check back and see how you’ve progressed.
I’ve finally gotten around to having a closer look at the issue you ran into but as per:
I have unfortunately been unable to reproduce the multi device system pool your description and pic showed; and the original reporter indicated. Could your fill in any more details as to how exactly you ended up with multi device system (rockstor_rockstor currently) pool. No worries otherwise, I expect in time we will get another report.
Sure I booted from an 8GB thumbdrive with the install image on it… My server is an ancient ASUS P7F in good condition. When I got tot he disk layout, it asked me about freeing space on my dual 1tb internal disks, then on the 32gb flash disk that was interneded to be my boot disk. The installer seemed to imply that he OS will be installed on the thumb disk, then then ended up creating one massive pool…
Hope that helps! I’m looking for a NAS solution to replace my aging wintel server fleet.
That’s exactly what I was after, thanks. It would now seem the issue you encountered concerns our current kickstart config and it’s interplay with your disk choices within the upstream (re-branded in our case) Anaconda installer.
There are currently tentative but active plans and ongoing efforts that will effectively address this issue in time so thanks for helping to clear up at least the location of this bug.
Best advise currently, with any installs / re-installs, is to detach all proposed or actual data drives prior to installing. That way the potential for accidental overwrite (with os) or of running into the same issue you reported, is minimised. Then once the new system is booted and configured our own Web-UI can take care of the data drive imports or wipes ready for use.
Thanks again for you help on this one and hope your adventures with Rockstor continue more smoothly.
