Note: This bug is displayed in read-only format because
the product is no longer active in Red Hat Bugzilla.
RHEL Engineering is moving the tracking of its product development work on RHEL 6 through RHEL 9 to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "RHEL project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs in the statuses "NEW", "ASSIGNED", and "POST" are being migrated throughout September 2023. Bugs of Red Hat partners with an assigned Engineering Partner Manager (EPM) are migrated in late September as per pre-agreed dates. Bugs against components "kernel", "kernel-rt", and "kpatch" are only migrated if still in "NEW" or "ASSIGNED". If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "RHEL project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/RHEL-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.
Update libxfs to take advantage of the efficient zeroing interface for block devices (REQ_OP_WRITE_ZEROES exposed via fallocate(blkdev_fd, FALLOC_FL_ZERO_RANGE) added in 2016-2017) instead of manually zeroing. This is particularly useful for network block devices such as rbd because manual zeroing with repeated (p)write() calls sends blocks of zeros over the network. scsi and nvme have various "write zeros" commands as well.
mkfs.ext4 was updated in v1.45.0. A quick test on a 10G rbd image shows a 6x improvement: 3.7 seconds with v1.44.4 vs 0.6 seconds with v1.45.3 (lazy_itable_init=1 by default, so it's just zeroing of the journal -- exactly what mkfs.xfs would do). For mkfs.xfs I expect the improvement to be smaller because xfs log is typically a lot smaller than ext4 journal (in this case it's 10M vs 64M for a 10G filesystem), but still worth it!
mkfs performance has become an important factor in the context of OCS (OpenShift Container Storage). Filesystem PVs (persistent volumes) are currently implemented as rbd images with a local filesystem on top. PV creation/deletion speeds is one of the key benchmarks. Creating and mapping an rbd image to a block device is almost instantaneous, so it boils down to mkfs performance.
I extracted the patch from the rpm and tested on top of v5.2.1 (just because that is what I've been testing with before). Works as expected:
10G image (10M log): 1.3 seconds vs 2.1 seconds
100G image (50M log): 1.5 seconds vs 5.2 seconds
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 (xfsprogs bug fix and enhancement update), 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-2020:4554