Description of problem: ----------------------- While choosing the disktype as 'raid5' doesn't set the proper chunksize for the thinpool Version-Release number of selected component (if applicable): ------------------------------------------------------------- gdeploy-2.0.2-4.el7rhgs How reproducible: ----------------- Always Steps to Reproduce: ------------------- 1. Create PV, VG, thinpool, LV using gdeploy config file with disktype as 'raid5', diskcount & stripesize Actual results: --------------- chunksize for the thinpool is not set to correct value Expected results: ----------------- chunksize for the thinpool should be set to the correct value
Thanks sac for finding out the issue in first place and also root causing the problem that code path computes chunksize only for raid6 disktype and not any other disktype <code_snip> if disktype == 'raid6': # If user has ommited chunk size or stripe unit size, we leave # it blank and let the system determine proper default value. if stripe_unit_size.strip() == '' or diskcount.strip() == '': return '' try: chunksize = str(int(stripe_unit_size) * int(diskcount)) + 'K' except ValueError: print "Only integer value is supported for stripesize or \ diskcount!" print "Found %s and %s"%(stripe_unit_size, diskcount) self.cleanup_and_quit() else: </code_snip>
Commit: https://github.com/gluster/gdeploy/commit/2bd7dc7 should fix the issue for RAID5
Tested with gdeploy-2.0.2-10.el7rhgs. With 'disktype' as 'raid5', 'chunksize' for the thinpool is calculated automatically to the correct value which is the product of 'stripesize' and 'diskcount'
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2017:2777