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the product is no longer active in Red Hat Bugzilla.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.
Description of problem:
Currently, it is documented that running VDO in sync mode on top of storage which requires flush/FUA for persistence is not a supported configuration. However, even if VDO is correctly started in sync mode on top of synchronous storage, it is possible for the write policy of the underlying storage to change without VDO being made aware. Hence, it is better for VDO to do what other components (such as XFS) do, and always issue sufficient flushes and/or FUAs to ensure correct behavior.
Version-Release number of selected component (if applicable):
How reproducible:
Demonstrating this issue is difficult, but the problem is always lurking.
Steps to Reproduce:
It is hard to quantify a reproduction of this issue, but:
1. Set up VDO in sync mode on top of storage with a write-back-cache enabled.
2. While writing data to the VDO, crash the system.
3. Observe more data loss than expected (this is the hard part, knowing what to expect).
Actual results:
Writes which were acknowledged before the crash did not persist after the crash.
Expected results:
All writes acknowledged before the crash persist after the crash.
Regression tests passed, but I tried to reproduce it to be sure:
I set up vdo in sync mode on writeback cache on iscsi disk. I created xfs fs on the vdo and mounted it. Thank I started to write 100 files to vdo, roughly around half through I removed the iscsi backing device and rebooted. With new version I get "cp: failed to access ‘/root/vdo_mount/random100.file’: Input/output error" on each file that is not acknowledged in user console.
With previous version I saw messages in /var/log/messages, see attachment 1401282[details]. I could not reproduce these messages (the 'spam') with the new version.
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/RHEA-2018:0900